LRDE
https://www.lrde.epita.fr/wiki/Special:Ask/-3Cq-3E-5B-5BCategory:News-7C-7CLrdeBulletin-5D-5D-20OR-20-5B-5BCategory:Publications-5D-5D-20-5B-5BPublished-20has-20news::true-5D-5D-20OR-20-5B-5BCategory:OlenaNews-5D-5D-20-5B-5BIs-20global-20news::true-5D-5D-3C-2Fq-3E/mainlabel%3D/limit%3D50/order%3D-20descending/sort%3D-20News-20date/offset%3D0/format%3D-20feed/searchlabel%3D-20-3Csmall-3ERSS-3C-2Fsmall-3E/page%3D-20full
From LRDEenMediaWiki 1.35.3Thu, 26 May 2022 15:14:15 GMTNewsEntry (2022/05/30)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/05/30)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/05/30)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2022/05/30)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>International conference <a rel="nofollow" class="external text" href="https://www.lix.polytechnique.fr/~smimram/getco22/">GETCO 2022</a> co-organized with École polytechnique at EPITA from 30 May to 3 June
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>GETCO is a conference series on Geometric and Topological Methods in Computer Science. The initial focus of GETCO was on concurrent and distributed computing, but the application area keeps expanding and now also includes higher categories and rewriting, dynamic and hybrid systems, robotics, and topological data analysis.
</td></tr>
<tr>
<th>Date
</th>
<td>2022/05/30
</td></tr></tbody></table>
</div>Wed, 25 May 2022 17:22:31 GMTDanielaNewsEntry (2022/05/23)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/05/23)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/05/23)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2022/05/23)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>Alexandre Duret-Lutz invited to Faculty of Informatics, Masaryk University, Czech Republic, for a week.
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>He will give a talk as part of the joint seminar <a rel="nofollow" class="external text" href="https://www.fi.muni.cz/dfseminar/index.html.en">of the DIMEA and FORMELA teams</a> on practical applications of the "Alternating Cycle Decomposition". His stay will be an opportunity to work on site with Jan Strejček, associate professor from FI MU.
</td></tr>
<tr>
<th>Date
</th>
<td>2022/05/23
</td></tr></tbody></table>
</div>Thu, 19 May 2022 14:29:20 GMTDanielaSome equivalence relation between persistent homology and morphological dynamics
https://www.lrde.epita.fr/wiki/Publications/boutry.22.jmiv.2
https://www.lrde.epita.fr/wiki/Publications/boutry.22.jmiv.2<div class="mw-parser-output"><p><a class="mw-selflink selflink">Some equivalence relation between persistent homology and morphological dynamics</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, Laurent Najman, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a></dd>
<dt>Journal</dt>
<dd>Journal of Mathematical Imaging and Vision</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2022-05-17</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>In Mathematical Morphology (MM), connected filters based on dynamics are used to filter the extrema of an image. Similarly, persistence is a concept coming from Persistent Homology (PH) and Morse Theory (MT) that represents the stability of the extrema of a Morse function. Since these two concepts seem to be closely related, in this paper we examine their relationship, and we prove that they are equal on <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a601995d55609f2d9f5e233e36fbe9ea26011b3b" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.395ex; height:1.676ex;" alt="{\displaystyle n}"/></span>-D Morse functions, <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n\geq 1}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
<mo>≥<!-- ≥ --></mo>
<mn>1</mn>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n\geq 1}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/d8ce9ce38d06f6bf5a3fe063118c09c2b6202bfe" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.505ex; width:5.656ex; height:2.343ex;" alt="{\displaystyle n\geq 1}"/></span>. More exactlypairing a minimum with a <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 1}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mn>1</mn>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle 1}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/92d98b82a3778f043108d4e20960a9193df57cbf" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.162ex; height:2.176ex;" alt="{\displaystyle 1}"/></span>-saddle by dynamics or pairing the same <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 1}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mn>1</mn>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle 1}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/92d98b82a3778f043108d4e20960a9193df57cbf" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.162ex; height:2.176ex;" alt="{\displaystyle 1}"/></span>-saddle with a minimum by persistence leads exactly to the same pairing, assuming that the critical values of the studied Morse function are unique. This result is a step further to show how much topological data analysis and mathematical morphology are relatedpaving the way for a more in-depth study of the relations between these two research fields.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/boutry.22.jmiv.2.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ boutry.22.jmiv.2,
author = {Nicolas Boutry and Laurent Najman and Thierry G\'eraud},
title = {Some equivalence relation between persistent homology and
morphological dynamics},
journal = {Journal of Mathematical Imaging and Vision},
volume = {},
number = {},
pages = {},
month = may,
year = {2022},
abstract = {In Mathematical Morphology (MM), connected filters based
on dynamics are used to filter the extrema of an image.
Similarly, persistence is a concept coming from Persistent
Homology (PH) and Morse Theory (MT) that represents the
stability of the extrema of a Morse function. Since these
two concepts seem to be closely related, in this paper we
examine their relationship, and we prove that they are
equal on $n$-D Morse functions, $n\geq 1$. More exactly,
pairing a minimum with a $1$-saddle by dynamics or pairing
the same $1$-saddle with a minimum by persistence leads
exactly to the same pairing, assuming that the critical
values of the studied Morse function are unique. This
result is a step further to show how much topological data
analysis and mathematical morphology are related, paving
the way for a more in-depth study of the relations between
these two research fields.},
doi = {}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 19 May 2022 03:58:58 GMTBotNewsEntry (2022/05/10)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/05/10)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/05/10)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2022/05/10)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>Uli Fahrenberg defends his <a href="/wiki/Affiche-these-HDR-UF" title="Affiche-these-HDR-UF"> Habilitation thesis</a> at Université Paris Saclay at 9 am.
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>
</td></tr>
<tr>
<th>Date
</th>
<td>2022/05/10
</td></tr></tbody></table>
</div>Mon, 23 May 2022 09:28:21 GMTDanielaLTL under reductions with weaker conditions than stutter invariance
https://www.lrde.epita.fr/wiki/Publications/paviot.22.forte
https://www.lrde.epita.fr/wiki/Publications/paviot.22.forte<div class="mw-parser-output"><p><a class="mw-selflink selflink">LTL under reductions with weaker conditions than stutter invariance</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Emmanuel Paviot-Adet, Denis Poitrenaud, <a href="/wiki/User:Renault" title="User:Renault">Etienne Renault</a>, Yann Thierry-Mieg</dd>
<dt>Where</dt>
<dd>Proceedings of the 41th IFIP International Conference on Formal Techniques for Distributed Objects, Components and Systems (FORTE'22)</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer&action=edit&redlink=1" class="new" title="Springer (page does not exist)">Springer</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Spot" title="Spot">Spot</a></dd>
<dt>Date</dt>
<dd>2022-04-18</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Verification of properties expressed as omega-regular languages such as LTL can benefit hugely from stutter insensitivity, using a diverse set of reduction strategies. However properties that are not stutter invariant, for instance due to the use of the neXt operator of LTL or to some form of counting in the logic, are not covered by these techniques in general. We propose in this paper to study a weaker property than stutter insensitivity. In a stutter insensitive language both adding and removing stutter to a word does not change its acceptance, any stuttering can be abstracted away; by decomposing this equivalence relation into two implications we obtain weaker conditions. We define a shortening insensitive language where any word that stutters less than a word in the language must also belong to the language. A lengthening insensitive language has the dual property. A semi-decision procedure is then introduced to reliably prove shortening insensitive properties or deny lengthening insensitive properties while working with a reduction of a system. A reduction has the property that it can only shorten runs. Lipton's transaction reductions or Petri net agglomerations are examples of eligible structural reduction strategies. An implementation and experimental evidence is provided showing most non- random properties sensitive to stutter are actually shortening or lengthening in- sensitive. Performance of experiments on a large (random) benchmark from the model-checking competition indicate that despite being a semi-decision proce- dure, the approach can still improve state of the art verification tools. 1 Introduction Model checking is an automatic verification technique for proving the correctness of systems that have finite state abstractions. Properties can be expressed using the popular Linear-time Temporal Logic (LTL). To verify LTL properties, the automata-theoretic approach [25] builds a product between a Buchi automaton representing the negation of the LTL formula and the reachable state graph of the system (seen as a set of infinite runs). This approach has been used successfully to verify both hardware and software components, but it suffers from the so called "state explosion problem": as the number of state variables in the system increases, the size of the system state space grows exponentially.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/paviot.22.forte.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ paviot.22.forte,
author = {Emmanuel Paviot-Adet and Denis Poitrenaud and Etienne
Renault and Yann Thierry-Mieg},
title = {LTL under reductions with weaker conditions than stutter
invariance},
booktitle = {Proceedings of the 41th IFIP International Conference on
Formal Techniques for Distributed Objects, Components and
Systems (FORTE'22)},
year = 2022,
month = jun,
series = {Lecture Notes in Computer Science},
volume = ??,
pages = {??--??},
publisher = {Springer},
doi = {??},
abstract = { Verification of properties expressed as omega-regular
languages such as LTL can benefit hugely from stutter
insensitivity, using a diverse set of reduction strategies.
However properties that are not stutter invariant, for
instance due to the use of the neXt operator of LTL or to
some form of counting in the logic, are not covered by
these techniques in general. We propose in this paper to
study a weaker property than stutter insensitivity. In a
stutter insensitive language both adding and removing
stutter to a word does not change its acceptance, any
stuttering can be abstracted away; by decomposing this
equivalence relation into two implications we obtain weaker
conditions. We define a shortening insensitive language
where any word that stutters less than a word in the
language must also belong to the language. A lengthening
insensitive language has the dual property. A semi-decision
procedure is then introduced to reliably prove shortening
insensitive properties or deny lengthening insensitive
properties while working with a reduction of a system. A
reduction has the property that it can only shorten runs.
Lipton's transaction reductions or Petri net agglomerations
are examples of eligible structural reduction strategies.
An implementation and experimental evidence is provided
showing most non- random properties sensitive to stutter
are actually shortening or lengthening in- sensitive.
Performance of experiments on a large (random) benchmark
from the model-checking competition indicate that despite
being a semi-decision proce- dure, the approach can still
improve state of the art verification tools. 1 Introduction
Model checking is an automatic verification technique for
proving the correctness of systems that have finite state
abstractions. Properties can be expressed using the popular
Linear-time Temporal Logic (LTL). To verify LTL properties,
the automata-theoretic approach [25] builds a product
between a Buchi automaton representing the negation of the
LTL formula and the reachable state graph of the system
(seen as a set of infinite runs). This approach has been
used successfully to verify both hardware and software
components, but it suffers from the so called "state
explosion problem": as the number of state variables in the
system increases, the size of the system state space grows
exponentially.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Mon, 16 May 2022 12:17:51 GMTBotEstimation of the noise level function for color images using mathematical morphology and non-parametric statistics
https://www.lrde.epita.fr/wiki/Publications/esteban.22.icpr
https://www.lrde.epita.fr/wiki/Publications/esteban.22.icpr<div class="mw-parser-output"><p><a class="mw-selflink selflink">Estimation of the noise level function for color images using mathematical morphology and non-parametric statistics</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Baptiste Esteban, <a href="/wiki/User:Gtochon" title="User:Gtochon">Guillaume Tochon</a>, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a>, <a href="/wiki/User:Didier" title="User:Didier">Didier Verna</a></dd>
<dt>Where</dt>
<dd>Proceedings of the 26th International Conference on Pattern Recognition</dd>
<dt>Place</dt>
<dd>Montréal, Québec</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2022-04-08</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Noise level information is crucial for many image processing tasks, such as image denoising. To estimate itit is necessary to find homegeneous areas within the image which contain only noise. Rank-based methods have proven to be efficient to achieve such a task. In the past, we proposed a method to estimate the noise level function (NLF) of grayscale images using the tree of shapes (ToS). This method, relying on the connected components extracted from the ToS computed on the noisy image, had the advantage of being adapted to the image content, which is not the case when using square blocks, but is still restricted to grayscale images. In this paper, we extend our ToS-based method to color images. Unlike grayscale images, the pixel values in multivariate images do not have a natural order relationship, which is a well-known issue when working with mathematical morphology and rank statistics. We propose to use the multivariate ToS to retrieve homogeneous regions. We derive an order relationship for the multivariate pixel values thanks to a complete lattice learning strategy and use it to compute the rank statistics. The obtained multivariate NLF is composed of one NLF per channel. The performance of the proposed method is compared with the one obtained using square blocks, and validates the soundness of the multivariate ToS structure for this task.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ esteban.22.icpr,
author = {Baptiste Esteban and Guillaume Tochon and Edwin Carlinet
and Didier Verna},
title = {Estimation of the noise level function for color images
using mathematical morphology and non-parametric
statistics},
booktitle = {Proceedings of the 26th International Conference on
Pattern Recognition},
year = 2022,
address = {Montr\'eal, Qu\'ebec},
category = {national},
month = aug,
abstract = {Noise level information is crucial for many image
processing tasks, such as image denoising. To estimate it,
it is necessary to find homegeneous areas within the image
which contain only noise. Rank-based methods have proven to
be efficient to achieve such a task. In the past, we
proposed a method to estimate the noise level function
(NLF) of grayscale images using the tree of shapes (ToS).
This method, relying on the connected components extracted
from the ToS computed on the noisy image, had the advantage
of being adapted to the image content, which is not the
case when using square blocks, but is still restricted to
grayscale images. In this paper, we extend our ToS-based
method to color images. Unlike grayscale images, the pixel
values in multivariate images do not have a natural order
relationship, which is a well-known issue when working with
mathematical morphology and rank statistics. We propose to
use the multivariate ToS to retrieve homogeneous regions.
We derive an order relationship for the multivariate pixel
values thanks to a complete lattice learning strategy and
use it to compute the rank statistics. The obtained
multivariate NLF is composed of one NLF per channel. The
performance of the proposed method is compared with the one
obtained using square blocks, and validates the soundness
of the multivariate ToS structure for this task. },
note = {accepted}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Tue, 26 Apr 2022 12:26:34 GMTBotA Benchmark of Named Entity Recognition Approaches in Historical Documents
https://www.lrde.epita.fr/wiki/Publications/abadie.22.das
https://www.lrde.epita.fr/wiki/Publications/abadie.22.das<div class="mw-parser-output"><p><a class="mw-selflink selflink">A Benchmark of Named Entity Recognition Approaches in Historical Documents</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Nathalie Abadie, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a>, <a href="/wiki/User:Chazalon" title="User:Chazalon">Joseph Chazalon</a>, Bertrand Duménieu</dd>
<dt>Where</dt>
<dd>Proceedings of the 15th IAPR International Workshop on Document Analysis System</dd>
<dt>Place</dt>
<dd>La Rochelle, France</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2022-04-07</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Named entity recognition (NER) is a necessary step in many pipelines targeting historical documents. Indeed, such natural language processing techniques identify which class each text token belongs to, e.g. “person name”“location”, “number”. Introducing a new public dataset built from 19th century French directories, we first assess how noisy modern, off-the-shelf OCR are. Then, we compare modern CNN- and Transformer-based NER techniques which can be reasonably used in the context of historical document analysis. We measure their requirements in terms of training data, the effects of OCR noise on their performance, and show how Transformer-based NER can benefit from unsupervised pre-training and supervised fine-tuning on noisy data. Results can be reproduced using resources available at <a rel="nofollow" class="external free" href="https://github.com/soduco/paper-ner-bench-das22">https://github.com/soduco/paper-ner-bench-das22</a> and <a rel="nofollow" class="external free" href="https://zenodo.org/record/6394464">https://zenodo.org/record/6394464</a>
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ abadie.22.das,
author = {Nathalie Abadie and Edwin Carlinet and Joseph Chazalon and
Bertrand Dum\'enieu},
title = {A Benchmark of Named Entity Recognition Approaches in
Historical Documents},
booktitle = {Proceedings of the 15th IAPR International Workshop on
Document Analysis System},
year = 2022,
address = {La Rochelle, France},
month = may,
abstract = {Named entity recognition (NER) is a necessary step in many
pipelines targeting historical documents. Indeed, such
natural language processing techniques identify which class
each text token belongs to, e.g. ``person name'',
``location'', ``number''. Introducing a new public dataset
built from 19th century French directories, we first assess
how noisy modern, off-the-shelf OCR are. Then, we compare
modern CNN- and Transformer-based NER techniques which can
be reasonably used in the context of historical document
analysis. We measure their requirements in terms of
training data, the effects of OCR noise on their
performance, and show how Transformer-based NER can benefit
from unsupervised pre-training and supervised fine-tuning
on noisy data. Results can be reproduced using resources
available at
https://github.com/soduco/paper-ner-bench-das22 and
https://zenodo.org/record/6394464},
note = {accepted}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Tue, 26 Apr 2022 12:26:01 GMTBotLearning Grayscale Mathematical Morphology with Smooth Morphological Layers
https://www.lrde.epita.fr/wiki/Publications/hermary.22.jmiv
https://www.lrde.epita.fr/wiki/Publications/hermary.22.jmiv<div class="mw-parser-output"><p><a class="mw-selflink selflink">Learning Grayscale Mathematical Morphology with Smooth Morphological Layers</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Romain Hermary, <a href="/wiki/User:Gtochon" title="User:Gtochon">Guillaume Tochon</a>, <a href="/wiki/User:Elodie" title="User:Elodie">Élodie Puybareau</a>, Alexandre Kirszenberg, Jesús Angulo</dd>
<dt>Journal</dt>
<dd>Journal of Mathematical Imaging and Vision</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2022-04-04</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>The integration of mathematical morphology operations within convolutional neural network architectures has received an increasing attention lately. Howeverreplacing standard convolution layers by morphological layers performing erosions or dilations is particularly challenging because the min and max operations are not differentiable. P-convolution layers were proposed as a possible solution to this issue since they can act as smooth differentiable approximation of min and max operations, yielding pseudo-dilation or pseudo-erosion layers. In a recent work, we proposed two novel morphological layers based on the same principle as the p-convolution, while circumventing its principal drawbacks, and showcased their capacity to efficiently learn grayscale morphological operators while raising several edge cases. In this work, we complete those previous results by thoroughly analyzing the behavior of the proposed layers and by investigating and settling the reported edge cases. We also demonstrate the compatibility of one of the proposed morphological layers with binary morphological frameworks.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/hermary.22.jmiv.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ hermary.22.jmiv,
author = {Romain Hermary and Guillaume Tochon and \'Elodie Puybareau
and Alexandre Kirszenberg and Jes\'us Angulo},
title = {Learning Grayscale Mathematical Morphology with Smooth
Morphological Layers},
journal = {Journal of Mathematical Imaging and Vision},
volume = {},
number = {},
pages = {},
month = apr,
year = {2022},
doi = {10.1007/s10851-022-01091-1},
abstract = {The integration of mathematical morphology operations
within convolutional neural network architectures has
received an increasing attention lately. However, replacing
standard convolution layers by morphological layers
performing erosions or dilations is particularly
challenging because the min and max operations are not
differentiable. P-convolution layers were proposed as a
possible solution to this issue since they can act as
smooth differentiable approximation of min and max
operations, yielding pseudo-dilation or pseudo-erosion
layers. In a recent work, we proposed two novel
morphological layers based on the same principle as the
p-convolution, while circumventing its principal drawbacks,
and showcased their capacity to efficiently learn grayscale
morphological operators while raising several edge cases.
In this work, we complete those previous results by
thoroughly analyzing the behavior of the proposed layers
and by investigating and settling the reported edge cases.
We also demonstrate the compatibility of one of the
proposed morphological layers with binary morphological
frameworks.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Mon, 16 May 2022 09:42:59 GMTBotQu'est-ce que mon GNN capture vraiment ? Exploration des représentations internes d'un GNN
https://www.lrde.epita.fr/wiki/Publications/veyrin-forrer.22.egc
https://www.lrde.epita.fr/wiki/Publications/veyrin-forrer.22.egc<div class="mw-parser-output"><p><a class="mw-selflink selflink">Qu'est-ce que mon GNN capture vraiment ? Exploration des représentations internes d'un GNN</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Luca Veyrin-Forrer, Ataollah Kamal, Stefan Duffner, Marc Plantevit, Céline Robardet</dd>
<dt>Where</dt>
<dd>Extraction et Gestion des Connaissances, EGC 2022Blois, France, 24 au 28 janvier 2022</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Keywords</dt>
<dd>IA</dd>
<dt>Date</dt>
<dd>2022-03-24</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>While existing GNN's explanation methods explain the decision by studying the output layer, we propose a method that analyzes the hidden layers to identify the neurons that are co-activated for a class. We associate to them a graph.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ veyrin-forrer.22.egc,
author = {Luca {Veyrin-Forrer} and Ataollah Kamal and Stefan Duffner
and Marc Plantevit and C\'eline Robardet},
title = {Qu'est-ce que mon {GNN} capture vraiment ? {E}xploration
des repr\'esentations internes d'un {GNN}},
booktitle = {Extraction et Gestion des Connaissances, {EGC} 2022,
Blois, France, 24 au 28 janvier 2022},
pages = {159--170},
year = {2022},
opturl = {http://editions-rnti.fr/?inprocid=1002725},
note = {In French, Best paper award},
category = {national},
abstract = {While existing GNN's explanation methods explain the
decision by studying the output layer, we propose a method
that analyzes the hidden layers to identify the neurons
that are co-activated for a class. We associate to them a
graph.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Tue, 05 Apr 2022 12:44:29 GMTBotLocal Intensity Order Transformation for Robust Curvilinear Object Segmentation
https://www.lrde.epita.fr/wiki/Publications/shi.21.itip
https://www.lrde.epita.fr/wiki/Publications/shi.21.itip<div class="mw-parser-output"><p><a class="mw-selflink selflink">Local Intensity Order Transformation for Robust Curvilinear Object Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Tianyi Shi, <a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, <a href="/wiki/User:Xu" title="User:Xu">Yongchao Xu</a>, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a></dd>
<dt>Journal</dt>
<dd>IEEE Transactions on Image Processing</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2022-03-22</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Segmentation of curvilinear structures is important in many applications, such as retinal blood vessel segmentation for early detection of vessel diseases and pavement crack segmentation for road condition evaluation and maintenance. Currently, deep learning-based methods have achieved impressive performance on these tasks. Yetmost of them mainly focus on finding powerful deep architectures but ignore capturing the inherent curvilinear structure feature (e.g., the curvilinear structure is darker than the context) for a more robust representation. In consequence, the performance usually drops a lot on cross-datasets, which poses great challenges in practice. In this paper, we aim to improve the generalizability by introducing a novel local intensity order transformation (LIOT). Specifically, we transfer a gray-scale image into a contrast- invariant four-channel image based on the intensity order between each pixel and its nearby pixels along with the four (horizontal and vertical) directions. This results in a representation that preserves the inherent characteristic of the curvilinear structure while being robust to contrast changes. Cross-dataset evaluation on three retinal blood vessel segmentation datasets demonstrates that LIOT improves the generalizability of some state-of-the-art methods. Additionally, the cross-dataset evaluation between retinal blood vessel segmentation and pavement crack segmentation shows that LIOT is able to preserve the inherent characteristic of curvilinear structure with large appearance gaps. An implementation of the proposed method is available at <a rel="nofollow" class="external free" href="https://github.com/TY-Shi/LIOT">https://github.com/TY-Shi/LIOT</a>.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/shi.22.itip.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ shi.21.itip,
author = {Tianyi Shi and Nicolas Boutry and Yongchao Xu and Thierry
G\'eraud},
title = {Local Intensity Order Transformation for Robust
Curvilinear Object Segmentation},
journal = {IEEE Transactions on Image Processing},
year = {2022},
volume = {31},
pages = {2557--2569},
month = mar,
abstract = {Segmentation of curvilinear structures is important in
many applications, such as retinal blood vessel
segmentation for early detection of vessel diseases and
pavement crack segmentation for road condition evaluation
and maintenance. Currently, deep learning-based methods
have achieved impressive performance on these tasks. Yet,
most of them mainly focus on finding powerful deep
architectures but ignore capturing the inherent curvilinear
structure feature (e.g., the curvilinear structure is
darker than the context) for a more robust representation.
In consequence, the performance usually drops a lot on
cross-datasets, which poses great challenges in practice.
In this paper, we aim to improve the generalizability by
introducing a novel local intensity order transformation
(LIOT). Specifically, we transfer a gray-scale image into a
contrast- invariant four-channel image based on the
intensity order between each pixel and its nearby pixels
along with the four (horizontal and vertical) directions.
This results in a representation that preserves the
inherent characteristic of the curvilinear structure while
being robust to contrast changes. Cross-dataset evaluation
on three retinal blood vessel segmentation datasets
demonstrates that LIOT improves the generalizability of
some state-of-the-art methods. Additionally, the
cross-dataset evaluation between retinal blood vessel
segmentation and pavement crack segmentation shows that
LIOT is able to preserve the inherent characteristic of
curvilinear structure with large appearance gaps. An
implementation of the proposed method is available at
\url{https://github.com/TY-Shi/LIOT}.},
url = {10.1109/TIP.2022.3155954}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Tue, 05 Apr 2022 12:44:22 GMTBotElectricity Price Forecasting on the Day-Ahead Market using Machine Learning
https://www.lrde.epita.fr/wiki/Publications/tschora.22.apen
https://www.lrde.epita.fr/wiki/Publications/tschora.22.apen<div class="mw-parser-output"><p><a class="mw-selflink selflink">Electricity Price Forecasting on the Day-Ahead Market using Machine Learning</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Léonard Tschora, Erwan Pierre, Marc Plantevit, Céline Robardet</dd>
<dt>Journal</dt>
<dd>Applied Energy</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Keywords</dt>
<dd>IA</dd>
<dt>Date</dt>
<dd>2022-03-10</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>The price of electricity on the European market is very volatile. This is due both to its mode of production by different sources, each with its own constraints (volume of production, dependence on the weather, or production inertia), and by the difficulty of its storage. Being able to predict the prices of the next day is an important issue, to allow the development of intelligent uses of electricity. In this article, we investigate the capabilities of different machine learning techniques to accurately predict electricity prices. Specifically, we extend current state-of-the-art approaches by considering previously unused predictive features such as price histories of neighboring countries. We show that these features significantly improve the quality of forecastseven in the current period when sudden changes are occurring. We also develop an analysis of the contribution of the different features in model prediction using Shap values, in order to shed light on how models make their prediction and to build user confidence in models.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/tschora.22.apen.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ tschora.22.apen,
author = {L\'eonard Tschora and Erwan Pierre and Marc Plantevit and
C\'eline Robardet},
title = {Electricity Price Forecasting on the Day-Ahead Market
using Machine Learning},
journal = {Applied Energy},
volume = {313},
number = {118752},
year = {2022},
doi = {10.1016/j.apenergy.2022.118752},
keywords = {Electricity price forecasting, Machine learning, Forecast
evaluation, Open-access benchmark, Explainable AI (XAI)},
abstract = {The price of electricity on the European market is very
volatile. This is due both to its mode of production by
different sources, each with its own constraints (volume of
production, dependence on the weather, or production
inertia), and by the difficulty of its storage. Being able
to predict the prices of the next day is an important
issue, to allow the development of intelligent uses of
electricity. In this article, we investigate the
capabilities of different machine learning techniques to
accurately predict electricity prices. Specifically, we
extend current state-of-the-art approaches by considering
previously unused predictive features such as price
histories of neighboring countries. We show that these
features significantly improve the quality of forecasts,
even in the current period when sudden changes are
occurring. We also develop an analysis of the contribution
of the different features in model prediction using Shap
values, in order to shed light on how models make their
prediction and to build user confidence in models.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 10 Mar 2022 13:45:04 GMTBotMax-Tree Computation on GPUs
https://www.lrde.epita.fr/wiki/Publications/blin.22.tpds
https://www.lrde.epita.fr/wiki/Publications/blin.22.tpds<div class="mw-parser-output"><p><a class="mw-selflink selflink">Max-Tree Computation on GPUs</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Nicolas Blin, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a>, Florian Lemaitre, Lionel Lacassagne, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a></dd>
<dt>Journal</dt>
<dd>IEEE Transactions on Parallel and Distributed Systems</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2022-03-09</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>In Mathematical Morphology, the max-tree is a region-based representation that encodes the inclusion relationship of the threshold sets of an image. This tree has been proven useful in numerous image processing applications. For the last decade, works have been led to improve the building time of this structure; mixing algorithmic optimizationsparallel and distributed computing. Nevertheless, there is still no algorithm that takes benefit from the computing power of the massively parallel architectures. In this work, we propose the first GPU algorithm to compute the max-tree. The proposed approach leads to significant speed-ups, and is up to one order of magnitude faster than the current State-of-the-Art parallel CPU algorithms. This work paves the way for a max-tree integration in image processing GPU pipelines and real-time image processing based on Mathematical Morphology. It is also a foundation for porting other image representations from Mathematical Morphology on GPUs.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/blin.22.tpds.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ blin.22.tpds,
author = {Nicolas Blin and Edwin Carlinet and Florian Lemaitre and
Lionel Lacassagne and Thierry G\'eraud},
title = {Max-Tree Computation on {GPU}s},
journal = {IEEE Transactions on Parallel and Distributed Systems},
month = mar,
year = {2022},
abstract = {In Mathematical Morphology, the max-tree is a region-based
representation that encodes the inclusion relationship of
the threshold sets of an image. This tree has been proven
useful in numerous image processing applications. For the
last decade, works have been led to improve the building
time of this structure; mixing algorithmic optimizations,
parallel and distributed computing. Nevertheless, there is
still no algorithm that takes benefit from the computing
power of the massively parallel architectures. In this
work, we propose the first GPU algorithm to compute the
max-tree. The proposed approach leads to significant
speed-ups, and is up to one order of magnitude faster than
the current State-of-the-Art parallel CPU algorithms. This
work paves the way for a max-tree integration in image
processing GPU pipelines and real-time image processing
based on Mathematical Morphology. It is also a foundation
for porting other image representations from Mathematical
Morphology on GPUs.},
doi = {10.1109/TPDS.2022.3158488}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 20 Apr 2022 07:02:41 GMTBotETAP: Experimental Typesetting Algorithms Platform
https://www.lrde.epita.fr/wiki/Publications/verna.22.els
https://www.lrde.epita.fr/wiki/Publications/verna.22.els<div class="mw-parser-output"><p><a class="mw-selflink selflink">ETAP: Experimental Typesetting Algorithms Platform</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Didier" title="User:Didier">Didier Verna</a></dd>
<dt>Where</dt>
<dd>15th European Lisp Symposium</dd>
<dt>Place</dt>
<dd>Porto, Portugal</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Date</dt>
<dd>2022-03-01</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>We present the early development stages of ETAP, a platform for experimenting with typesetting algorithms. The purpose of this platform is twofold: while its primary objective is to provide building blocks for quickly and easily designing and testing new algorithms (or variations on existing ones), it can also be used as an interactivereal time demonstrator for many features of digital typography, such as kerning, hyphenation, or ligaturing.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ verna.22.els,
author = {Didier Verna},
title = {{ETAP}: Experimental Typesetting Algorithms Platform},
booktitle = {15th European Lisp Symposium},
year = 2022,
month = mar,
address = {Porto, Portugal},
isbn = 9782955747469,
doi = {10.5281/zenodo.6334248},
abstract = {We present the early development stages of ETAP, a
platform for experimenting with typesetting algorithms. The
purpose of this platform is twofold: while its primary
objective is to provide building blocks for quickly and
easily designing and testing new algorithms (or variations
on existing ones), it can also be used as an interactive,
real time demonstrator for many features of digital
typography, such as kerning, hyphenation, or ligaturing.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Tue, 05 Apr 2022 12:44:29 GMTBotPractical Applications of the Alternating Cycle Decomposition
https://www.lrde.epita.fr/wiki/Publications/casares.22.tacas
https://www.lrde.epita.fr/wiki/Publications/casares.22.tacas<div class="mw-parser-output"><p><a class="mw-selflink selflink">Practical Applications of the Alternating Cycle Decomposition</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Antonio Casares, <a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/~adl/">Alexandre Duret-Lutz</a>, Klara J Meyer, <a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/~frenkin/">Florian Renkin</a>, Salomon Sickert</dd>
<dt>Where</dt>
<dd>Proceedings of the 28th International Conference on Tools and Algorithms for the Construction and Analysis of Systems</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Spot" title="Spot">Spot</a></dd>
<dt>Date</dt>
<dd>2022-02-01</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>In 2021, Casares, Colcombet, and Fijalkow introduced the Alternating Cycle Decomposition (ACD) to study properties and transformations of Muller automata. We present the first practical implementation of the ACD in two different tools, Owl and Spot, and adapt it to the framework of Emerson-Lei automata, i.e., <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \omega }">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>ω<!-- ω --></mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle \omega }</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/48eff443f9de7a985bb94ca3bde20813ea737be8" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.446ex; height:1.676ex;" alt="{\displaystyle \omega }"/></span>-automata whose acceptance conditions are defined by Boolean formulas. The ACD provides a transformation of Emerson-Lei automata into parity automata with strong optimality guarantees: the resulting parity automaton is minimal among those automata that can be obtained by duplication of states. Our empirical results show that this transformation is usable in practice. Further, we show how the ACD can generalize many other specialized constructions such as deciding typeness of automata and degeneralization of generalized Büchi automata, providing a framework of practical algorithms for <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \omega }">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>ω<!-- ω --></mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle \omega }</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/48eff443f9de7a985bb94ca3bde20813ea737be8" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.446ex; height:1.676ex;" alt="{\displaystyle \omega }"/></span>-automata.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/casares.22.tacas.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ casares.22.tacas,
author = {Antonio Casares and Alexandre Duret-Lutz and Klara J.
Meyer and Florian Renkin and Salomon Sickert},
title = {Practical Applications of the {A}lternating {C}ycle
{D}ecomposition},
booktitle = {Proceedings of the 28th International Conference on Tools
and Algorithms for the Construction and Analysis of
Systems},
year = {2022},
series = {Lecture Notes in Computer Science},
volume = {13244},
month = apr,
abstract = {In 2021, Casares, Colcombet, and Fijalkow introduced the
Alternating Cycle Decomposition (ACD) to study properties
and transformations of Muller automata. We present the
first practical implementation of the ACD in two different
tools, Owl and Spot, and adapt it to the framework of
Emerson-Lei automata, i.e., $\omega$-automata whose
acceptance conditions are defined by Boolean formulas. The
ACD provides a transformation of Emerson-Lei automata into
parity automata with strong optimality guarantees: the
resulting parity automaton is minimal among those automata
that can be obtained by duplication of states. Our
empirical results show that this transformation is usable
in practice. Further, we show how the ACD can generalize
many other specialized constructions such as deciding
typeness of automata and degeneralization of generalized
B{\"u}chi automata, providing a framework of practical
algorithms for $\omega$-automata.},
pages = {99--117},
doi = {10.1007/978-3-030-99527-0_6}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Tue, 05 Apr 2022 12:43:45 GMTBotNewsEntry (2022/01/01)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/01/01)
https://www.lrde.epita.fr/wiki/NewsEntry_(2022/01/01)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2022/01/01)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>LRDE is happy to welcome a new member, Marc Plantevit
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>Marc Plantevit holds a PhD in Computer Science from <a rel="nofollow" class="external text" href="https://www.umontpellier.fr">Montpellier University</a> and a HDR from <a rel="nofollow" class="external text" href="https://www.universite-lyon.fr">Lyon University</a>. Before joining EPITA Lyon and LRDE, he was an associate professor at <a rel="nofollow" class="external text" href="https://www.univ-lyon1.fr">University Claude Bernard Lyon 1</a> and head of the <a rel="nofollow" class="external text" href="https://liris.cnrs.fr/en/team/dm2l">Data Mining & Machine Learning research group</a> at LIRIS lab. His research is mainly concerned with foundation of data mining, graph mining, subgroup discovery and explainable artificial intelligence. He is also interested in the application of machine learning in wide applications such as Neuroscience (olfaction), Electricity price forecasting, recommender systems, etc.
</td></tr>
<tr>
<th>Date
</th>
<td>2022/01/01
</td></tr></tbody></table>
</div>Tue, 08 Feb 2022 07:51:08 GMTDanielaAGAT: Building and Evaluating Binary Partition Trees for Image Segmentation
https://www.lrde.epita.fr/wiki/Publications/randrianasoa.21.softx
https://www.lrde.epita.fr/wiki/Publications/randrianasoa.21.softx<div class="mw-parser-output"><p><a class="mw-selflink selflink">AGAT: Building and Evaluating Binary Partition Trees for Image Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Jimmy Francky Randrianasoa, Camille Kurtz, Éric Desjardin, Nicolas Passat</dd>
<dt>Journal</dt>
<dd>SoftwareX</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Elsevier&action=edit&redlink=1" class="new" title="Elsevier (page does not exist)">Elsevier</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Date</dt>
<dd>2021-12-17</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>AGAT is a Java library dedicated to the constructionhandling and evaluation of binary partition trees, a hierarchical data structure providing multiscale partitioning of images and, more generally, of valued graphs. On the one hand, this library offers functionalities to build binary partition trees in the usual way, but also to define multifeature trees, a novel and richer paradigm of binary partition trees built from multiple images and/or several criteria. On the other handit also allows one to manipulate the binary partition trees, for instance by defining/computing tree cuts that can be interpreted in particular as segmentations when dealing with image modeling. In addition, some evaluation tools are also provided, which allow one to evaluate the quality of different binary partition trees for such segmentation tasks. AGAT can be easily handled by various kinds of users (students, researchers, practitioners). It can be used as is for experimental purposes, but can also form a basis for the development of new methods and paradigms for construction, use and intensive evaluation of binary partition trees. Beyond the usual imaging applications, its underlying structure also allows for more general developments in graph-based analysis, leading to a wide range of potential applications in computer visionimage/data analysis and machine learning.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ randrianasoa.21.softx,
author = {Jimmy Francky Randrianasoa and Camille Kurtz and \'Eric
Desjardin and Nicolas Passat},
title = {{AGAT}: {B}uilding and Evaluating Binary Partition Trees
for Image Segmentation},
journal = {SoftwareX},
year = 2021,
volume = {16},
number = {100855},
month = dec,
publisher = {Elsevier},
abstract = {AGAT is a Java library dedicated to the construction,
handling and evaluation of binary partition trees, a
hierarchical data structure providing multiscale
partitioning of images and, more generally, of valued
graphs. On the one hand, this library offers
functionalities to build binary partition trees in the
usual way, but also to define multifeature trees, a novel
and richer paradigm of binary partition trees built from
multiple images and/or several criteria. On the other hand,
it also allows one to manipulate the binary partition
trees, for instance by defining/computing tree cuts that
can be interpreted in particular as segmentations when
dealing with image modeling. In addition, some evaluation
tools are also provided, which allow one to evaluate the
quality of different binary partition trees for such
segmentation tasks. AGAT can be easily handled by various
kinds of users (students, researchers, practitioners). It
can be used as is for experimental purposes, but can also
form a basis for the development of new methods and
paradigms for construction, use and intensive evaluation of
binary partition trees. Beyond the usual imaging
applications, its underlying structure also allows for more
general developments in graph-based analysis, leading to a
wide range of potential applications in computer vision,
image/data analysis and machine learning.},
doi = {10.1016/j.softx.2021.100855}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 10 Mar 2022 13:44:59 GMTBotIntroducing the Boundary-Aware Loss for Deep Image Segmentation
https://www.lrde.epita.fr/wiki/Publications/movn.21.bmvc
https://www.lrde.epita.fr/wiki/Publications/movn.21.bmvc<div class="mw-parser-output"><p><a class="mw-selflink selflink">Introducing the Boundary-Aware Loss for Deep Image Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Movn" title="User:Movn">Minh Ôn Vũ Ngoc</a>, Yizi Chen, <a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, <a href="/wiki/User:Chazalon" title="User:Chazalon">Joseph Chazalon</a>, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a>, <a href="/wiki/User:Jonathan" title="User:Jonathan">Jonathan Fabrizio</a>, Clément Mallet, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a></dd>
<dt>Where</dt>
<dd>Proceedings of the 32nd British Machine Vision Conference (BMVC)</dd>
<dt>Place</dt>
<dd>Online</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-11-28</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Most contemporary supervised image segmentation methods do not preserve the initial topology of the given input (like the closeness of the contours). One can generally remark that edge points have been inserted or removed when the binary prediction and the ground truth are compared. This can be critical when accurate localization of multiple interconnected objects is required. In this paper, we present a new loss function, called, Boundary-Aware loss (BALoss), based on the Minimum Barrier Distance (MBD) cut algorithm. It is able to locate what we call the it leakage pixels and to encode the boundary information coming from the given ground truth. Thanks to this adapted loss, we are able to significantly refine the quality of the predicted boundaries during the learning procedure. Furthermore, our loss function is differentiable and can be applied to any kind of neural network used in image processing. We apply this loss function on the standard U-Net and DC U-Net on Electron Microscopy datasets. They are well-known to be challenging due to their high noise level and to the close or even connected objects covering the image space. Our segmentation performance, in terms of Variation of Information (VOI) and Adapted Rank Index (ARI), are very promising and lead to <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \approx {}15\%}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mo>≈<!-- ≈ --></mo>
<mrow class="MJX-TeXAtom-ORD">
</mrow>
<mn>15</mn>
<mi mathvariant="normal">%<!-- % --></mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle \approx {}15\%}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/8ef0f92119ded1eae1be4e30276c0fe1ffbd955b" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:6.714ex; height:2.343ex;" alt="{\displaystyle \approx {}15\%}"/></span> better scores of VOI and <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \approx {}5\%}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mo>≈<!-- ≈ --></mo>
<mrow class="MJX-TeXAtom-ORD">
</mrow>
<mn>5</mn>
<mi mathvariant="normal">%<!-- % --></mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle \approx {}5\%}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/c49fefef45fdc230183405078b511495ca21e438" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:5.552ex; height:2.343ex;" alt="{\displaystyle \approx {}5\%}"/></span> better scores of ARI than the state-of-the-art. The code of boundary-awareness loss is freely available at <a rel="nofollow" class="external free" href="https://github.com/onvungocminh/MBD_BAL">https://github.com/onvungocminh/MBD_BAL</a>
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/movn.21.bmvc.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ movn.21.bmvc,
author = {Minh \^On V\~{u} Ng\d{o}c and Yizi Chen and Nicolas Boutry
and Joseph Chazalon and Edwin Carlinet and Jonathan
Fabrizio and Cl\'ement Mallet and Thierry G\'eraud},
title = {Introducing the Boundary-Aware Loss for Deep Image
Segmentation},
booktitle = {Proceedings of the 32nd British Machine Vision Conference
(BMVC)},
year = 2021,
address = {Online},
abstract = {Most contemporary supervised image segmentation methods do
not preserve the initial topology of the given input (like
the closeness of the contours). One can generally remark
that edge points have been inserted or removed when the
binary prediction and the ground truth are compared. This
can be critical when accurate localization of multiple
interconnected objects is required. In this paper, we
present a new loss function, called, Boundary-Aware loss
(BALoss), based on the Minimum Barrier Distance (MBD) cut
algorithm. It is able to locate what we call the {\it
leakage pixels} and to encode the boundary information
coming from the given ground truth. Thanks to this adapted
loss, we are able to significantly refine the quality of
the predicted boundaries during the learning procedure.
Furthermore, our loss function is differentiable and can be
applied to any kind of neural network used in image
processing. We apply this loss function on the standard
U-Net and DC U-Net on Electron Microscopy datasets. They
are well-known to be challenging due to their high noise
level and to the close or even connected objects covering
the image space. Our segmentation performance, in terms of
Variation of Information (VOI) and Adapted Rank Index
(ARI), are very promising and lead to $\approx{}15\%$
better scores of VOI and $\approx{}5\%$ better scores of
ARI than the state-of-the-art. The code of
boundary-awareness loss is freely available at
\url{https://github.com/onvungocminh/MBD_BAL}},
note = {https://www.bmvc2021-virtualconference.com/assets/papers/1546.pdf}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 19 May 2022 03:59:38 GMTBotStrong Euler Wellcomposedness
https://www.lrde.epita.fr/wiki/Publications/boutry.21.joco
https://www.lrde.epita.fr/wiki/Publications/boutry.21.joco<div class="mw-parser-output"><p><a class="mw-selflink selflink">Strong Euler Wellcomposedness</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, Rocio Gonzalez-Diaz, Maria-Jose Jimenez, Eduardo Paluzo-Hildago</dd>
<dt>Journal</dt>
<dd>Journal of Combinatorial Optimization</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-11-23</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>In this paper, we define a new flavour of well-composedness, called strong Euler well-composedness. In the general setting of regular cell complexes, a regular cell complex of dimension <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a601995d55609f2d9f5e233e36fbe9ea26011b3b" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.395ex; height:1.676ex;" alt="{\displaystyle n}"/></span> is strongly Euler well-composed if the Euler characteristic of the link of each boundary cell is <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 1}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mn>1</mn>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle 1}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/92d98b82a3778f043108d4e20960a9193df57cbf" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.162ex; height:2.176ex;" alt="{\displaystyle 1}"/></span>, which is the Euler characteristic of an <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle (n-1)}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mo stretchy="false">(</mo>
<mi>n</mi>
<mo>−<!-- − --></mo>
<mn>1</mn>
<mo stretchy="false">)</mo>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle (n-1)}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/df88c6333caaf6471cf277f24b802ff9931b133e" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.838ex; width:7.207ex; height:2.843ex;" alt="{\displaystyle (n-1)}"/></span>-dimensional ball. Working in the particular setting of cubical complexes canonically associated with <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a601995d55609f2d9f5e233e36fbe9ea26011b3b" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.395ex; height:1.676ex;" alt="{\displaystyle n}"/></span>-D pictures, we formally prove in this paper that strong Euler well-composedness implies digital well-composedness in any dimension <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n\geq 2}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
<mo>≥<!-- ≥ --></mo>
<mn>2</mn>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n\geq 2}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/e6bf67f9d06ca3af619657f8d20ee1322da77174" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.505ex; width:5.656ex; height:2.343ex;" alt="{\displaystyle n\geq 2}"/></span> and that the converse is not true when <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n\geq 4}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
<mo>≥<!-- ≥ --></mo>
<mn>4</mn>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n\geq 4}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/25010fec4b0f68f1b46f49d14917d962acca0b16" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.505ex; width:5.656ex; height:2.343ex;" alt="{\displaystyle n\geq 4}"/></span>.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/boutry.21.joco.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ boutry.21.joco,
author = {Nicolas Boutry and Rocio Gonzalez-Diaz and Maria-Jose
Jimenez and Eduardo Paluzo-Hildago},
title = {Strong {E}uler Wellcomposedness},
journal = {Journal of Combinatorial Optimization},
month = nov,
year = {2021},
abstract = {In this paper, we define a new flavour of
well-composedness, called strong Euler well-composedness.
In the general setting of regular cell complexes, a regular
cell complex of dimension $n$ is strongly Euler
well-composed if the Euler characteristic of the link of
each boundary cell is $1$, which is the Euler
characteristic of an $(n-1)$-dimensional ball. Working in
the particular setting of cubical complexes canonically
associated with $n$-D pictures, we formally prove in this
paper that strong Euler well-composedness implies digital
well-composedness in any dimension $n\geq 2$ and that the
converse is not true when $n\geq 4$.},
doi = {10.1007/s10878-021-00837-8}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 10 Mar 2022 13:44:14 GMTBotContinuous Well-Composedness implies Digital Well-Composedness in n-D
https://www.lrde.epita.fr/wiki/Publications/boutry.22.jmiv
https://www.lrde.epita.fr/wiki/Publications/boutry.22.jmiv<div class="mw-parser-output"><p><a class="mw-selflink selflink">Continuous Well-Composedness implies Digital Well-Composedness in n-D</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, Rocio Gonzalez-Diaz, Laurent Najman, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a></dd>
<dt>Journal</dt>
<dd>Journal of Mathematical Imaging and Vision</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-11-09</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>In this paper, we prove that when a <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a601995d55609f2d9f5e233e36fbe9ea26011b3b" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.395ex; height:1.676ex;" alt="{\displaystyle n}"/></span>-D cubical set is continuously well-composed (CWC), that is, when the boundary of its continuous analog is a topological <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle (n-1)}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mo stretchy="false">(</mo>
<mi>n</mi>
<mo>−<!-- − --></mo>
<mn>1</mn>
<mo stretchy="false">)</mo>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle (n-1)}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/df88c6333caaf6471cf277f24b802ff9931b133e" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.838ex; width:7.207ex; height:2.843ex;" alt="{\displaystyle (n-1)}"/></span>-manifold, then it is digitally well-composed (DWC)which means that it does not contain any critical configuration. We prove this result thanks to local homology. This paper is the sequel of a previous paper where we proved that DWCness does not imply CWCness in 4D.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/boutry.22.jmiv.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ boutry.22.jmiv,
author = {Nicolas Boutry and Rocio Gonzalez-Diaz and Laurent Najman
and Thierry G\'eraud},
title = {Continuous Well-Composedness implies Digital
Well-Composedness in $n$-D},
journal = {Journal of Mathematical Imaging and Vision},
volume = {64},
number = {2},
pages = {131--150},
month = jan,
year = {2022},
abstract = {In this paper, we prove that when a $n$-D cubical set is
continuously well-composed (CWC), that is, when the
boundary of its continuous analog is a topological
$(n-1)$-manifold, then it is digitally well-composed (DWC),
which means that it does not contain any critical
configuration. We prove this result thanks to local
homology. This paper is the sequel of a previous paper
where we proved that DWCness does not imply CWCness in
4D.},
doi = {10.1007/s10851-021-01058-8}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 10 Mar 2022 13:44:15 GMTBotNewsEntry (2021/11/02)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/11/02)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/11/02)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2021/11/02)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>Another PhD student at LRDE, Thibault Buatois, who joins the <a href="/wiki/Olena" title="Olena">Olena</a> team.
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>Holding <a rel="nofollow" class="external text" href="https://www.epita.fr/nos-formations/diplome-ingenieur/cycle-ingenieur/les-majeures/">EPITA's degree with IMAGE and RDI double major</a>, Thibault comes back to LRDE for a PhD in medical imaging. He will continue to work on medical image segmentation, using lightweight neural networks and medical knowledge as well as adding explainability to the segmentation process.
</td></tr>
<tr>
<th>Date
</th>
<td>2021/11/02
</td></tr></tbody></table>
</div>Thu, 18 Nov 2021 17:18:51 GMTDanielaNewsEntry (2021/10/18)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/10/18)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/10/18)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2021/10/18)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>The LRDE hosts a new PhD student, Antoine Martin, who joins the <a href="/wiki/Spot" title="Spot">Spot</a> team.
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>After completing <a rel="nofollow" class="external text" href="https://www.epita.fr/nos-formations/diplome-ingenieur/cycle-ingenieur/les-majeures/">EPITA's IMAGE and RDI double major</a>, Antoine is back at LRDE for his PhD. Having worked on parallel algorithms for automata and model checking, then on a model checker for Go programs, he will now focus on efficient translation of industrial temporal logics to ω-automata.
</td></tr>
<tr>
<th>Date
</th>
<td>2021/10/18
</td></tr></tbody></table>
</div>Thu, 18 Nov 2021 17:22:15 GMTDanielaSeminar/2021-10-06
https://www.lrde.epita.fr/wiki/Seminar/2021-10-06
https://www.lrde.epita.fr/wiki/Seminar/2021-10-06<div class="mw-parser-output"><p><a class="mw-selflink selflink">Seminar/2021-10-06</a>
</p></div><div class="mw-parser-output"><h3><span id="Mercredi_6_octobre_2021,_11h_-_12h,_Https://meet.jit.si/SeminaireLRDE_\&_IP_13"></span><span class="mw-headline" id="Mercredi_6_octobre_2021.2C_11h_-_12h.2C_Https:.2F.2Fmeet.jit.si.2FSeminaireLRDE_.5C.26_IP_13"><a class="mw-selflink selflink"> Mercredi 6 octobre 2021, 11h - 12h, Https://meet.jit.si/SeminaireLRDE \& IP 13</a></span></h3>
<p><br />
</p>
<h4><span class="mw-headline" id="Scaling_Optimal_Transport_for_High_Dimensional_Learning">Scaling Optimal Transport for High Dimensional Learning</span></h4>
<p><i>Gabriel Peyré, CNRS and Ecole Normale Supérieure</i>
<br />
<br />
</p><p>Optimal transport (OT) has recently gained a lot of interest in machine learning. It is a natural tool to compare in a geometrically faithful way probability distributions. It finds applications in both supervised learning (using geometric loss functions) and unsupervised learning (to perform generative model fitting). OT is however plagued by the curse of dimensionality, since it might require a number of samples which grows exponentially with the dimension. In this talk, I will review entropic regularization methods which define geometric loss functions approximating OT with a better sample complexity.
<br />
<br />
</p><p><small>Gabriel Peyré is a CNRS senior researcher and professor at Ecole Normale Supérieure, Paris. He works at the interface between applied mathematics, imaging and machine learning. He obtained 2 ERC grants (Starting in 2010 and Consolidator in 2017), the Blaise Pascal prize from the French academy of sciences in 2017, the Magenes Prize from the Italian Mathematical Union in 2019 and the silver medal from CNRS in 2021. He is invited speaker at the European Congress for Mathematics in 2020. He is the deputy director of the Prairie Institute for artificial intelligence, the director of the ENS center for data science and the former director of the GdR CNRS MIA. He is the head of the ELLIS (European Lab for Learning & Intelligent Systems) Paris Unit. He is engaged in reproducible research and code education.</small>
<br />
<br />
<a rel="nofollow" class="external text" href="https://optimaltransport.github.io/">https://optimaltransport.github.io/</a>, <a rel="nofollow" class="external text" href="http://www.numerical-tours.com/">http://www.numerical-tours.com/</a>, <a rel="nofollow" class="external text" href="https://ellis-paris.github.io/">https://ellis-paris.github.io/</a>
</p></div>Tue, 28 Sep 2021 09:15:02 GMTBotNewsEntry (2021/09/02)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/09/02)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/09/02)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2021/09/02)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>LRDE’s <a href="/wiki/Spot" title="Spot">Spot</a> team is happy to welcome another member, Sven Dziadek, for a one-year postdoc.
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>Sven recently completed his PhD at <a rel="nofollow" class="external text" href="https://www.uni-leipzig.de/">Leipzig University</a> where he considered weighted ω-automata. Weighted automata are used to describe quantitative properties of systems. At LRDE, Sven will investigate quantitative model checking and contribute to <a href="/wiki/Spot" title="Spot">Spot</a>.
</td></tr>
<tr>
<th>Date
</th>
<td>2021/09/02
</td></tr></tbody></table>
</div>Mon, 15 Nov 2021 11:28:46 GMTDanielaNewsEntry (2021/09/01)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/09/01)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/09/01)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2021/09/01)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>LRDE's <a href="/wiki/Spot" title="Spot">Spot</a> team hosts a new member, Uli Fahrenberg
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>Ulrich (Uli) Fahrenberg holds a PhD in algebraic topology from <a rel="nofollow" class="external text" href="https://www.en.aau.dk">Aalborg University, Denmark</a>. After a postdoc at <a rel="nofollow" class="external text" href="https://www.inria.fr/fr/centre-inria-rennes-bretagne-atlantique">Inria Rennes</a>, followed by a position at <a rel="nofollow" class="external text" href="https://www.polytechnique.edu">École polytechnique</a>, he starts now as associate professor at EPITA Rennes and will work together with LRDE’s <a href="/wiki/Spot" title="Spot">Spot</a> team on automata theory, concurrency theory, real-time verification, and general quantitative verification.
</td></tr>
<tr>
<th>Date
</th>
<td>2021/09/01
</td></tr></tbody></table>
</div>Mon, 15 Nov 2021 11:31:03 GMTDanielaTowards better Heuristics for solving Bounded Model Checking Problems
https://www.lrde.epita.fr/wiki/Publications/kheireddine.21.cp
https://www.lrde.epita.fr/wiki/Publications/kheireddine.21.cp<div class="mw-parser-output"><p><a class="mw-selflink selflink">Towards better Heuristics for solving Bounded Model Checking Problems</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Anissa Kheireddine, <a href="/wiki/User:Renault" title="User:Renault">Étienne Renault</a>, Souheib Baarrir</dd>
<dt>Where</dt>
<dd>Proceedings of the 27th International Conference on Principles and Practice of Constraint Programmings (CP)</dd>
<dt>Place</dt>
<dd>Montpellier, France (Virtual Conference)</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Schloss_Dagstuhl_%E2%80%93_Leibniz-Zentrum_f%C3%BCr_Informatik&action=edit&redlink=1" class="new" title="Schloss Dagstuhl – Leibniz-Zentrum für Informatik (page does not exist)">Schloss Dagstuhl – Leibniz-Zentrum für Informatik</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Spot" title="Spot">Spot</a></dd>
<dt>Date</dt>
<dd>2021-08-31</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>This paper presents a new way to improve the performance of the SAT-based bounded model checking problem by exploiting relevant information identified through the characteristics of the original problem. This led us to design a new way of building interesting heuristics based on the structure of the underlying problem. The proposed methodology is generic and can be applied for any SAT problem. This paper compares the state-of-the-art approach with two new heuristics: Structure-based and Linear Programming heuristics and show promising results.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/kheireddine.21.cp.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ kheireddine.21.cp,
author = {Anissa Kheireddine and \'Etienne Renault and Souheib
Baarrir},
title = {Towards better Heuristics for solving Bounded Model
Checking Problems},
booktitle = {Proceedings of the 27th International Conference on
Principles and Practice of Constraint Programmings (CP)},
editor = {Laurent D. Michel},
series = {Leibniz International Proceedings in Informatics
(LIPIcs)},
address = {Montpellier, France (Virtual Conference)},
year = {2021},
pages = {7:1--7:11},
volume = {210},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
month = oct,
abstract = {This paper presents a new way to improve the performance
of the SAT-based bounded model checking problem by
exploiting relevant information identified through the
characteristics of the original problem. This led us to
design a new way of building interesting heuristics based
on the structure of the underlying problem. The proposed
methodology is generic and can be applied for any SAT
problem. This paper compares the state-of-the-art approach
with two new heuristics: Structure-based and Linear
Programming heuristics and show promising results.},
doi = {10.4230/LIPIcs.CP.2021.7},
isbn = {978-3-95977-211-2},
issn = {1868-8969}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 12 Jan 2022 08:29:43 GMTBotVerSe: A Vertebrae Labelling and Segmentation Benchmark for Multi-detector CT Images
https://www.lrde.epita.fr/wiki/Publications/sekuboyina.21.media
https://www.lrde.epita.fr/wiki/Publications/sekuboyina.21.media<div class="mw-parser-output"><p><a class="mw-selflink selflink">VerSe: A Vertebrae Labelling and Segmentation Benchmark for Multi-detector CT Images</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Anjany Sekuboyina, Malek E Husseini, Amirhossein Bayat, Maximilian Löffler, Hans Liebl, Hongwei Li, Giles Tetteh, Jan Kukačka, Christian Payer, Darko Stern, Martin Urschler, Maodong Chen, Dalong Cheng, Nikolas Lessmann, Yujin Hu, Tianfu Wang, Dong Yang, Daguang Xu, and Felix Ambellan, Tamaz Amiranashvili, Moritz Ehlke, Hans Lamecker, Sebastian Lehnert, Marilia Lirio, Nicolás Pérez de Olaguer, Heiko Ramm, Manish Sahu, Alexander Tack, Stefan Zachow, Tao Jiang, Xinjun Ma, Christoph Angerman, Xin Wang, Kevin Brown, Matthias Wolf, Alexandre Kirszenberg, <a href="/wiki/User:Elodie" title="User:Elodie">Élodie Puybareau</a>, Di Chen, Yiwei Bai, Brandon H Rapazzo, Timyoas Yeah, Amber Zhang, Shangliang Xu, Feng Houa, Zhiqiang He, Chan Zeng, Zheng Xiangshang, Xu Liming, Tucker J Netherton, Raymond P Mumme, Laurence E Court, Zixun Huang, Chenhang He, Li-Wen Wang, Sai Ho Ling, <a href="/wiki/User:Dhuynh" title="User:Dhuynh">Lê Duy Huỳnh</a>, <a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, Roman Jakubicek, Jiri Chmelik, Supriti Mulay, Mohanasankar Sivaprakasam, Johannes C Paetzold, Suprosanna Shit, Ivan Ezhov, Benedikt Wiestler, Ben Glocker, Alexander Valentinitsch, Markus Rempfler, Björn H Menze, Jan S Kirschke</dd>
<dt>Journal</dt>
<dd>Medical Image Analysis</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Date</dt>
<dd>2021-07-22</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision support systems for diagnosissurgery planning, and population-based analysis of spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms tackling the labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel level by a human-machine hybrid algorithm (<a rel="nofollow" class="external free" href="https://osf.io/nqjyw/">https://osf.io/nqjyw/</a>, urlhttps://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the results of this evaluation and further investigate the performance variation at the vertebra level, scan level, and different fields of view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top-performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The VerSe content and code can be accessed at: <a rel="nofollow" class="external free" href="https://github.com/anjany/verse">https://github.com/anjany/verse</a>.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/sekuboyina.21.media.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ sekuboyina.21.media,
author = {Anjany Sekuboyina and Malek E. Husseini and Amirhossein
Bayat and Maximilian L\"offler and Hans Liebl and Hongwei
Li and Giles Tetteh and Jan Kuka\v{c}ka and Christian Payer
and Darko Stern and Martin Urschler and Maodong Chen and
Dalong Cheng and Nikolas Lessmann and Yujin Hu and Tianfu
Wang and Dong Yang and Daguang Xu and and Felix Ambellan
and Tamaz Amiranashvili and Moritz Ehlke and Hans Lamecker
and Sebastian Lehnert and Marilia Lirio and Nicol\'as
{P\'erez de Olaguer} and Heiko Ramm and Manish Sahu and
Alexander Tack and Stefan Zachow and Tao Jiang and Xinjun
Ma and Christoph Angerman and Xin Wang and Kevin Brown and
Matthias Wolf and Alexandre Kirszenberg and \'Elodie
Puybareau and Di Chen and Yiwei Bai and Brandon H. Rapazzo
and Timyoas Yeah and Amber Zhang and Shangliang Xu and Feng
Houa and Zhiqiang He and Chan Zeng and Zheng Xiangshang and
Xu Liming and Tucker J. Netherton and Raymond P. Mumme and
Laurence E. Court and Zixun Huang and Chenhang He and
Li-Wen Wang and Sai Ho Ling and L\^e Duy Hu\`ynh and
Nicolas Boutry and Roman Jakubicek and Jiri Chmelik and
Supriti Mulay and Mohanasankar Sivaprakasam and Johannes C.
Paetzold and Suprosanna Shit and Ivan Ezhov and Benedikt
Wiestler and Ben Glocker and Alexander Valentinitsch and
Markus Rempfler and Bj\"orn H. Menze and Jan S. Kirschke},
title = {{VerSe}: {A} Vertebrae Labelling and Segmentation
Benchmark for Multi-detector {CT} Images},
journal = {Medical Image Analysis},
number = {102166},
year = {2021},
month = jul,
doi = {10.1016/j.media.2021.102166},
abstract = {Vertebral labelling and segmentation are two fundamental
tasks in an automated spine processing pipeline. Reliable
and accurate processing of spine images is expected to
benefit clinical decision support systems for diagnosis,
surgery planning, and population-based analysis of spine
and bone health. However, designing automated algorithms
for spine processing is challenging predominantly due to
considerable variations in anatomy and acquisition
protocols and due to a severe shortage of publicly
available data. Addressing these limitations, the Large
Scale Vertebrae Segmentation Challenge (VerSe) was
organised in conjunction with the International Conference
on Medical Image Computing and Computer Assisted
Intervention (MICCAI) in 2019 and 2020, with a call for
algorithms tackling the labelling and segmentation of
vertebrae. Two datasets containing a total of 374
multi-detector CT scans from 355 patients were prepared and
4505 vertebrae have individually been annotated at voxel
level by a human-machine hybrid algorithm
(\url{https://osf.io/nqjyw/}, \url{https://osf.io/t98fz/}).
A total of 25 algorithms were benchmarked on these
datasets. In this work, we present the results of this
evaluation and further investigate the performance
variation at the vertebra level, scan level, and different
fields of view. We also evaluate the generalisability of
the approaches to an implicit domain shift in data by
evaluating the top-performing algorithms of one challenge
iteration on data from the other iteration. The principal
takeaway from VerSe: the performance of an algorithm in
labelling and segmenting a spine scan hinges on its ability
to correctly identify vertebrae in cases of rare anatomical
variations. The VerSe content and code can be accessed at:
\url{https://github.com/anjany/verse}.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:58:09 GMTBotNewsEntry (2021/07/05)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/07/05)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/07/05)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2021/07/05)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>LRDE researcher Guillaume Tochon participates in LEMONADE project selected by ANR for a JCJC 2021 grant
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>The LEMONADE project (LEarning and MOdeliNg spectrAl Dynamics of satellite image time sEries) has been selected by the French National Research Agency as a research project coordinated by young researchers (JCJC). The project’s principal investigator is Lucas Drumetz (<a rel="nofollow" class="external text" href="https://www.imt-atlantique.fr">IMT-Atlantique, Lab-STICC</a>), with Mauro Dalla Mura (<a rel="nofollow" class="external text" href="http://www.gipsa-lab.fr">Grenoble-INP, GIPSA-Lab</a>) and Guillaume Tochon from LRDE as partners. The project will start in October 2021. The goal of this project is to learn and model, with deep neural network approaches, the spectral dynamics of satellite image time series.
</td></tr>
<tr>
<th>Date
</th>
<td>2021/07/05
</td></tr></tbody></table>
</div>Wed, 28 Jul 2021 12:53:00 GMTDanielaGo2Pins: A Framework for the LTL Verification of Go Programs
https://www.lrde.epita.fr/wiki/Publications/kirszenberg.21.spin
https://www.lrde.epita.fr/wiki/Publications/kirszenberg.21.spin<div class="mw-parser-output"><p><a class="mw-selflink selflink">Go2Pins: A Framework for the LTL Verification of Go Programs</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Alexandre Kirszenberg, Antoine Martin, Hugo Moreau, <a href="/wiki/User:Renault" title="User:Renault">Etienne Renault</a></dd>
<dt>Where</dt>
<dd>Proceedings of the 27th International SPIN Symposium on Model Checking of Software (SPIN'21)</dd>
<dt>Place</dt>
<dd>Aarhus, Denmark (online)</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer,_Cham&action=edit&redlink=1" class="new" title="Springer, Cham (page does not exist)">Springer, Cham</a></dd>
<dt>Keywords</dt>
<dd>Spot</dd>
<dt>Date</dt>
<dd>2021-06-08</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>We introduce Go2Pins, a tool that takes a program written in Go and links it with two model-checkers: LTSMin [19] and Spot [7]. Go2Pins is an effort to promote the integration of both formal verifica- tion and testing inside industrial-size projects. With this goal in mind, we introduce black-box transitions, an efficient and scalable technique for handling the Go runtime. This approachinspired by hardware ver- ification techniques, allows easy, automatic and efficient abstractions. Go2Pins also handles basic concurrent programs through the use of a dedicated scheduler. In this paper we demonstrate the usage of Go2Pins over benchmarks inspired by industrial problems and a set of LTL formulae. Even if Go2Pins is still at the early stages of development, our results are promising and show the the benefits of using black-box transitions.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/kirszenberg.21.spin.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ kirszenberg.21.spin,
author = {Alexandre Kirszenberg and Antoine Martin and Hugo Moreau
and Etienne Renault},
title = {{Go2Pins}: {A} Framework for the {LTL} Verification of
{Go} Programs},
booktitle = {Proceedings of the 27th International SPIN Symposium on
Model Checking of Software (SPIN'21)},
year = {2021},
series = {Lecture Notes in Computer Science},
volume = {12864},
month = may,
address = {Aarhus, Denmark (online)},
publisher = {Springer, Cham},
pages = {140--156},
abstract = {We introduce Go2Pins, a tool that takes a program written
in Go and links it with two model-checkers: LTSMin [19] and
Spot [7]. Go2Pins is an effort to promote the integration
of both formal verifica- tion and testing inside
industrial-size projects. With this goal in mind, we
introduce black-box transitions, an efficient and scalable
technique for handling the Go runtime. This approach,
inspired by hardware ver- ification techniques, allows
easy, automatic and efficient abstractions. Go2Pins also
handles basic concurrent programs through the use of a
dedicated scheduler. In this paper we demonstrate the usage
of Go2Pins over benchmarks inspired by industrial problems
and a set of LTL formulae. Even if Go2Pins is still at the
early stages of development, our results are promising and
show the the benefits of using black-box transitions.},
doi = {10.1007/978-3-030-84629-9_8}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Mon, 06 Sep 2021 13:06:46 GMTBotVectorization of Historical Maps Using Deep Edge Filtering and Closed Shape Extraction
https://www.lrde.epita.fr/wiki/Publications/chen.21.icdar
https://www.lrde.epita.fr/wiki/Publications/chen.21.icdar<div class="mw-parser-output"><p><a class="mw-selflink selflink">Vectorization of Historical Maps Using Deep Edge Filtering and Closed Shape Extraction</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Yizi Chen, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a>, <a href="/wiki/User:Chazalon" title="User:Chazalon">Joseph Chazalon</a>, Clément Mallet, Bertrand Duménieu, Julien Perret</dd>
<dt>Where</dt>
<dd>Proceedings of the 16th International Conference on Document Analysis and Recognition (ICDAR'21)</dd>
<dt>Place</dt>
<dd>Lausanne, Switzerland</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer,_Cham&action=edit&redlink=1" class="new" title="Springer, Cham (page does not exist)">Springer, Cham</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-05-17</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Maps have been a unique source of knowledge for centuries. Such historical documents provide invaluable information for analyzing the complex spatial transformation of landscapes over important time frames. This is particularly true for urban areas that encompass multiple interleaved research domains (social sciences, economy, etc.). The large amount and significant diversity of map sources call for automatic image processing techniques in order to extract the relevant objects under a vectorial shape. The complexity of maps (text, noise, digitization artifactsetc.) has hindered the capacity of proposing a versatile and efficient raster-to-vector approaches for decades. We propose a learnable, reproducible, and reusable solution for the automatic transformation of raster maps into vector objects (building blocks, streets, rivers). It is built upon the complementary strength of mathematical morphology and convolutional neural networks through efficient edge filtering. Evenmore, we modify ConnNet and combine with deep edge filtering architecture to make use of pixel connectivity information and built an end-to-end system without requiring any post-processing techniques. In this paper, we focus on the comprehensive benchmark on various architectures on multiple datasets coupled with a novel vectorization step. Our experimental results on a new public dataset using COCO Panoptic metric exhibit very encouraging results confirmed by a qualitative analysis of the success and failure cases of our approach. Codedataset, results and extra illustrations are freely available at <a rel="nofollow" class="external free" href="https://github.com/soduco/ICDAR-2021-Vectorization">https://github.com/soduco/ICDAR-2021-Vectorization</a>.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/chen.21.icdar.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ chen.21.icdar,
title = {Vectorization of Historical Maps Using Deep Edge Filtering
and Closed Shape Extraction},
author = {Yizi Chen and Edwin Carlinet and Joseph Chazalon and
Cl\'ement Mallet and Bertrand Dum\'enieu and Julien Perret},
booktitle = {Proceedings of the 16th International Conference on
Document Analysis and Recognition (ICDAR'21)},
year = {2021},
month = sep,
pages = {510--525},
series = {Lecture Notes in Computer Science},
publisher = {Springer, Cham},
volume = {12824},
address = {Lausanne, Switzerland},
abstract = {Maps have been a unique source of knowledge for centuries.
Such historical documents provide invaluable information
for analyzing the complex spatial transformation of
landscapes over important time frames. This is particularly
true for urban areas that encompass multiple interleaved
research domains (social sciences, economy, etc.). The
large amount and significant diversity of map sources call
for automatic image processing techniques in order to
extract the relevant objects under a vectorial shape. The
complexity of maps (text, noise, digitization artifacts,
etc.) has hindered the capacity of proposing a versatile
and efficient raster-to-vector approaches for decades. We
propose a learnable, reproducible, and reusable solution
for the automatic transformation of raster maps into vector
objects (building blocks, streets, rivers). It is built
upon the complementary strength of mathematical morphology
and convolutional neural networks through efficient edge
filtering. Evenmore, we modify ConnNet and combine with
deep edge filtering architecture to make use of pixel
connectivity information and built an end-to-end system
without requiring any post-processing techniques. In this
paper, we focus on the comprehensive benchmark on various
architectures on multiple datasets coupled with a novel
vectorization step. Our experimental results on a new
public dataset using COCO Panoptic metric exhibit very
encouraging results confirmed by a qualitative analysis of
the success and failure cases of our approach. Code,
dataset, results and extra illustrations are freely
available at
\url{https://github.com/soduco/ICDAR-2021-Vectorization}. },
doi = {10.1007/978-3-030-86337-1_34}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:55:11 GMTBotICDAR 2021 Competition on Historical Map Segmentation
https://www.lrde.epita.fr/wiki/Publications/chazalon.21.icdar.2
https://www.lrde.epita.fr/wiki/Publications/chazalon.21.icdar.2<div class="mw-parser-output"><p><a class="mw-selflink selflink">ICDAR 2021 Competition on Historical Map Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Chazalon" title="User:Chazalon">Joseph Chazalon</a>, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a>, Yizi Chen, Julien Perret, Bertrand Duménieu, Clément Mallet, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a>, Vincent Nguyen, Nam Nguyen, Josef Baloun, Ladislav Lenc, Pavel Král</dd>
<dt>Where</dt>
<dd>Proceedings of the 16th International Conference on Document Analysis and Recognition (ICDAR'21)</dd>
<dt>Place</dt>
<dd>Lausanne, Switzerland</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer,_Cham&action=edit&redlink=1" class="new" title="Springer, Cham (page does not exist)">Springer, Cham</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-05-17</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>This paper presents the final results of the ICDAR 2021 Competition on Historical Map Segmentation (MapSeg)encouraging research on a series of historical atlases of Paris, France, drawn at 1/5000 scale between 1894 and 1937. The competition featured three tasks, awarded separately. Task 1 consists in detecting building blocks and was won by the L3IRIS team using a DenseNet-121 network trained in a weakly supervised fashion. This task is evaluated on 3 large images containing hundreds of shapes to detect. Task 2 consists in segmenting map content from the larger map sheet, and was won by the UWB team using a U-Net-like FCN combined with a binarization method to increase detection edge accuracy. Task 3 consists in locating intersection points of geo-referencing lines, and was also won by the UWB team who used a dedicated pipeline combining binarization, line detection with Hough transformcandidate filtering, and template matching for intersection refinement. Tasks 2 and 3 are evaluated on 95 map sheets with complex content. Dataset, evaluation tools and results are available under permissive licensing at <a rel="nofollow" class="external free" href="https://icdar21-mapseg.github.io/">https://icdar21-mapseg.github.io/</a>.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/chazalon.21.icdar.2.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ chazalon.21.icdar.2,
title = {{ICDAR} 2021 Competition on Historical Map Segmentation},
author = {Joseph Chazalon and Edwin Carlinet and Yizi Chen and
Julien Perret and Bertrand Dum\'enieu and Cl\'ement Mallet
and Thierry G\'eraud and Vincent Nguyen and Nam Nguyen and
Josef Baloun and Ladislav Lenc and Pavel Kr\'al},
booktitle = {Proceedings of the 16th International Conference on
Document Analysis and Recognition (ICDAR'21)},
year = {2021},
month = sep,
pages = {693--707},
series = {Lecture Notes in Computer Science},
publisher = {Springer, Cham},
volume = {12824},
address = {Lausanne, Switzerland},
abstract = {This paper presents the final results of the ICDAR 2021
Competition on Historical Map Segmentation (MapSeg),
encouraging research on a series of historical atlases of
Paris, France, drawn at 1/5000 scale between 1894 and 1937.
The competition featured three tasks, awarded separately.
Task~1 consists in detecting building blocks and was won by
the L3IRIS team using a DenseNet-121 network trained in a
weakly supervised fashion. This task is evaluated on 3
large images containing hundreds of shapes to detect.
Task~2 consists in segmenting map content from the larger
map sheet, and was won by the UWB team using a U-Net-like
FCN combined with a binarization method to increase
detection edge accuracy. Task~3 consists in locating
intersection points of geo-referencing lines, and was also
won by the UWB team who used a dedicated pipeline combining
binarization, line detection with Hough transform,
candidate filtering, and template matching for intersection
refinement. Tasks~2 and~3 are evaluated on 95 map sheets
with complex content. Dataset, evaluation tools and results
are available under permissive licensing at
\url{https://icdar21-mapseg.github.io/}.},
doi = {10.1007/978-3-030-86337-1_46}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:55:01 GMTBotRevisiting the Coco Panoptic Metric to Enable Visual and Qualitative Analysis of Historical Map Instance Segmentation
https://www.lrde.epita.fr/wiki/Publications/chazalon.21.icdar.1
https://www.lrde.epita.fr/wiki/Publications/chazalon.21.icdar.1<div class="mw-parser-output"><p><a class="mw-selflink selflink">Revisiting the Coco Panoptic Metric to Enable Visual and Qualitative Analysis of Historical Map Instance Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Chazalon" title="User:Chazalon">Joseph Chazalon</a>, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a></dd>
<dt>Where</dt>
<dd>Proceedings of the 16th International Conference on Document Analysis and Recognition (ICDAR'21)</dd>
<dt>Place</dt>
<dd>Lausanne, Switzerland</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer,_Cham&action=edit&redlink=1" class="new" title="Springer, Cham (page does not exist)">Springer, Cham</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-05-17</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Segmentation is an important task. It is so important that there exist tens of metrics trying to score and rank segmentation systems. It is so important that each topic has its own metric because their problem is too specific. Does it? What are the fundamental differences with the ZoneMap metric used for page segmentation, the COCO Panoptic metric used in computer vision and metrics used to rank hierarchical segmentations? In this paper, while assessing segmentation accuracy for historical maps, we explain, compare and demystify some the most used segmentation evaluation protocols. In particular, we focus on an alternative view of the COCO Panoptic metric as a classification evaluation; we show its soundness and propose extensions with more “shape-oriented” metrics. Beyond a quantitative metric, this paper aims also at providing qualitative measures through precision-recall maps that enable visualizing the success and the failures of a segmentation method.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/chazalon.21.icdar.1.pdf">Paper</a></li></ul>
<p><br />
</p>
<ul><li><a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/chazalon.21.icdar.1.poster.pdf">Poster</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ chazalon.21.icdar.1,
title = {Revisiting the {C}oco Panoptic Metric to Enable Visual and
Qualitative Analysis of Historical Map Instance
Segmentation},
author = {Joseph Chazalon and Edwin Carlinet},
booktitle = {Proceedings of the 16th International Conference on
Document Analysis and Recognition (ICDAR'21)},
year = {2021},
month = sep,
series = {Lecture Notes in Computer Science},
publisher = {Springer, Cham},
volume = {12824},
pages = {367--382},
address = {Lausanne, Switzerland},
abstract = {Segmentation is an important task. It is so important that
there exist tens of metrics trying to score and rank
segmentation systems. It is so important that each topic
has its own metric because their problem is too specific.
Does it? What are the fundamental differences with the
ZoneMap metric used for page segmentation, the COCO
Panoptic metric used in computer vision and metrics used to
rank hierarchical segmentations? In this paper, while
assessing segmentation accuracy for historical maps, we
explain, compare and demystify some the most used
segmentation evaluation protocols. In particular, we focus
on an alternative view of the COCO Panoptic metric as a
classification evaluation; we show its soundness and
propose extensions with more ``shape-oriented'' metrics.
Beyond a quantitative metric, this paper aims also at
providing qualitative measures through
\emph{precision-recall maps} that enable visualizing the
success and the failures of a segmentation method.},
doi = {10.1007/978-3-030-86337-1_25}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:54:57 GMTBotSeminar/2021-05-12
https://www.lrde.epita.fr/wiki/Seminar/2021-05-12
https://www.lrde.epita.fr/wiki/Seminar/2021-05-12<div class="mw-parser-output"><p><a class="mw-selflink selflink">Seminar/2021-05-12</a>
</p></div><div class="mw-parser-output"><h3><span id="Mercredi_12_mai_2021,_11h_-_12h,_Https://meet.jit.si/SeminaireLRDE"></span><span class="mw-headline" id="Mercredi_12_mai_2021.2C_11h_-_12h.2C_Https:.2F.2Fmeet.jit.si.2FSeminaireLRDE"><a class="mw-selflink selflink"> Mercredi 12 mai 2021, 11h - 12h, Https://meet.jit.si/SeminaireLRDE</a></span></h3>
<p><br />
</p>
<h4><span class="mw-headline" id="An_Introduction_to_Topological_Data_Analysis_with_the_Topology_ToolKit">An Introduction to Topological Data Analysis with the Topology ToolKit</span></h4>
<p><i>Julien Tierny, Sorbonne Université</i>
<br />
<br />
</p><p>Topological Data Analysis (TDA) is a recent area of computer science that focuses on discovering intrinsic structures hidden in data. Based on solid mathematical tools such as Morse theory and Persistent Homology, TDA enables the robust extraction of the main features of a data set into stable, concise, and multi-scale descriptors that facilitate data analysis and visualization. In this talk, I will give an intuitive overview of the main tools used in TDA (persistence diagrams, Reeb graphs, Morse-Smale complexes, etc.) with applications to concrete use cases in computational fluid dynamics, medical imaging, quantum chemistry, and climate modeling. This talk will be illustrated with results produced with the "Topology ToolKit" (TTK), an open-source library (BSD license) that we develop with collaborators to showcase our research. Tutorials for re-producing these experiments are available on the TTK website.
<br />
<br />
</p><p><small>Julien Tierny received his Ph.D. degree in Computer Science from the University of Lille in 2008 and
the Habilitation degree (HDR) from Sorbonne University in 2016. Currently a CNRS permanent
research scientist affiliated with Sorbonne University, his research expertise lies in topological methods
for data analysis and visualization. Author on the topic and award winner for his research, he regularly
serves as an international program committee member for the top venues in data visualization (IEEE VIS,
EuroVis, etc.) and is an associate editor for IEEE Transactions on Visualization and Computer Graphics.
Julien Tierny is also founder and lead developer of the Topology ToolKit (TTK), an open source library for
topological data analysis.</small>
<br />
<br />
<a rel="nofollow" class="external text" href="https://topology-tool-kit.github.io/">https://topology-tool-kit.github.io/</a>
</p></div>Wed, 28 Apr 2021 11:35:37 GMTBotLearning Sentinel-2 Spectral Dynamics for Long-Run Predictions Using Residual Neural Networks
https://www.lrde.epita.fr/wiki/Publications/estopinan.21.eusipco
https://www.lrde.epita.fr/wiki/Publications/estopinan.21.eusipco<div class="mw-parser-output"><p><a class="mw-selflink selflink">Learning Sentinel-2 Spectral Dynamics for Long-Run Predictions Using Residual Neural Networks</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Joaquim Estopinan, <a href="/wiki/User:Gtochon" title="User:Gtochon">Guillaume Tochon</a>, Lucas Drumetz</dd>
<dt>Where</dt>
<dd>Proceedings of the 29th European Signal Processing Conference (EUSIPCO)</dd>
<dt>Place</dt>
<dd>Dublin, Ireland</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-05-04</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Making the most of multispectral image time-series is a promising but still relatively under-explored research direction because of the complexity of jointly analyzing spatial, spectral and temporal information. Capturing and characterizing temporal dynamics is one of the important and challenging issues. Our new method paves the way to capture real data dynamics and should eventually benefit applications like unmixing or classification. Dealing with time-series dynamics classically requires the knowledge of a dynamical model and an observation model. The former may be incorrect or computationally hard to handle, thus motivating data-driven strategies aiming at learning dynamics directly from data. In this paper, we adapt neural network architectures to learn periodic dynamics of both simulated and real multispectral time-series. We emphasize the necessity of choosing the right state variable to capture periodic dynamics and show that our models can reproduce the average seasonal dynamics of vegetation using only one year of training data.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ estopinan.21.eusipco,
author = {Joaquim Estopinan and Guillaume Tochon and Lucas Drumetz},
title = {Learning {Sentinel-2} Spectral Dynamics for Long-Run
Predictions Using Residual Neural Networks},
booktitle = {Proceedings of the 29th European Signal Processing
Conference (EUSIPCO)},
year = 2021,
address = {Dublin, Ireland},
month = aug,
abstract = {Making the most of multispectral image time-series is a
promising but still relatively under-explored research
direction because of the complexity of jointly analyzing
spatial, spectral and temporal information. Capturing and
characterizing temporal dynamics is one of the important
and challenging issues. Our new method paves the way to
capture real data dynamics and should eventually benefit
applications like unmixing or classification. Dealing with
time-series dynamics classically requires the knowledge of
a dynamical model and an observation model. The former may
be incorrect or computationally hard to handle, thus
motivating data-driven strategies aiming at learning
dynamics directly from data. In this paper, we adapt neural
network architectures to learn periodic dynamics of both
simulated and real multispectral time-series. We emphasize
the necessity of choosing the right state variable to
capture periodic dynamics and show that our models can
reproduce the average seasonal dynamics of vegetation using
only one year of training data.},
doi = {10.23919/EUSIPCO54536.2021.9616304}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 12 Jan 2022 10:33:28 GMTBotA Corpus Processing and Analysis Pipeline for Quickref
https://www.lrde.epita.fr/wiki/Publications/hacquard.21.els
https://www.lrde.epita.fr/wiki/Publications/hacquard.21.els<div class="mw-parser-output"><p><a class="mw-selflink selflink">A Corpus Processing and Analysis Pipeline for Quickref</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Antoine Hacquard, <a href="/wiki/User:Didier" title="User:Didier">Didier Verna</a></dd>
<dt>Where</dt>
<dd>Proceedings of the 14th European Lisp Symposium (ELS)</dd>
<dt>Place</dt>
<dd>Online</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Date</dt>
<dd>2021-05-01</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Quicklisp is a library manager working with your existing Common Lisp implementation to download and install around 2000 libraries, from a central archive. Quickref, an application itself written in Common Lisp, generatesautomatically and by introspection, a technical documentation for every library in Quicklisp, and produces a website for this documentation. In this paper, we present a corpus processing and analysis pipeline for Quickref. This pipeline consists of a set of natural language processing blocks allowing us to analyze Quicklisp libraries, based on natural language contents sources such as README files, docstrings, or symbol names. The ultimate purpose of this pipeline is the generation of a keyword index for Quickref, although other applications such as word clouds or topic analysis are also envisioned.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ hacquard.21.els,
author = {Antoine Hacquard and Didier Verna},
title = {A Corpus Processing and Analysis Pipeline for {Q}uickref},
booktitle = {Proceedings of the 14th European Lisp Symposium (ELS)},
year = 2021,
pages = {27--35},
month = may,
address = {Online},
isbn = 9782955747452,
doi = {10.5281/zenodo.4714443},
abstract = {Quicklisp is a library manager working with your existing
Common Lisp implementation to download and install around
2000 libraries, from a central archive. Quickref, an
application itself written in Common Lisp, generates,
automatically and by introspection, a technical
documentation for every library in Quicklisp, and produces
a website for this documentation. In this paper, we present
a corpus processing and analysis pipeline for Quickref.
This pipeline consists of a set of natural language
processing blocks allowing us to analyze Quicklisp
libraries, based on natural language contents sources such
as README files, docstrings, or symbol names. The ultimate
purpose of this pipeline is the generation of a keyword
index for Quickref, although other applications such as
word clouds or topic analysis are also envisioned.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:56:33 GMTBotA Portable, Simple, Embeddable Type System
https://www.lrde.epita.fr/wiki/Publications/newton.21.els
https://www.lrde.epita.fr/wiki/Publications/newton.21.els<div class="mw-parser-output"><p><a class="mw-selflink selflink">A Portable, Simple, Embeddable Type System</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Jnewton" title="User:Jnewton">Jim Newton</a>, <a href="/wiki/User:Adrien" title="User:Adrien">Adrien Pommellet</a></dd>
<dt>Where</dt>
<dd>Proceedings of the 14th European Lisp Symposium (ELS)</dd>
<dt>Place</dt>
<dd>Online</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=European_Lisp_Symposium&action=edit&redlink=1" class="new" title="European Lisp Symposium (page does not exist)">European Lisp Symposium</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Spot" title="Spot">Spot</a></dd>
<dt>Keywords</dt>
<dd>infinite alphabets, type systems, Common Lisp, ClojureScala</dd>
<dt>Date</dt>
<dd>2021-04-26</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>We present a simple type system inspired by that of Common Lisp. The type system is intended to be embedded into a host language and accepts certain fundamental types from that language as axiomatically given. The type calculus provided in the type system is capable of expressing union, intersection, and complement types, as well as membership, subtype, disjoint, and habitation (non-emptiness) checks. We present a theoretical foundation and two sample implementations, one in Clojure and one in Scala.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/newton.21.els.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ newton.21.els,
author = {Jim Newton and Adrien Pommellet},
title = {A Portable, Simple, Embeddable Type System},
booktitle = {Proceedings of the 14th European Lisp Symposium (ELS)},
year = 2021,
lrdestatus = {accepted},
address = {Online},
month = may,
abstract = { We present a simple type system inspired by that of
Common Lisp. The type system is intended to be embedded
into a host language and accepts certain fundamental types
from that language as axiomatically given. The type
calculus provided in the type system is capable of
expressing union, intersection, and complement types, as
well as membership, subtype, disjoint, and habitation
(non-emptiness) checks. We present a theoretical foundation
and two sample implementations, one in Clojure and one in
Scala.},
pages = {11--20},
publisher = {European Lisp Symposium},
doi = {10.5281/zenodo.4709777}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Tue, 11 Jan 2022 14:56:40 GMTBotSeminar/2021-03-31
https://www.lrde.epita.fr/wiki/Seminar/2021-03-31
https://www.lrde.epita.fr/wiki/Seminar/2021-03-31<div class="mw-parser-output"><p><a class="mw-selflink selflink">Seminar/2021-03-31</a>
</p></div><div class="mw-parser-output"><h3><span id="Mercredi_31_mars_2021,_11h_-_12h,_Https://meet.jit.si/SeminaireLRDE_\&_Amphi_4"></span><span class="mw-headline" id="Mercredi_31_mars_2021.2C_11h_-_12h.2C_Https:.2F.2Fmeet.jit.si.2FSeminaireLRDE_.5C.26_Amphi_4"><a class="mw-selflink selflink"> Mercredi 31 mars 2021, 11h - 12h, Https://meet.jit.si/SeminaireLRDE \& Amphi 4</a></span></h3>
<p><br />
</p>
<h4><span class="mw-headline" id="Contributions_to_Boolean_satisfiability_solving_and_its_application_to_the_analysis_of_discrete_systems">Contributions to Boolean satisfiability solving and its application to the analysis of discrete systems</span></h4>
<p><i>Souheib Baarir, Université Paris VI</i>
<br />
<br />
</p><p>Despite its NP-completeness, propositional Boolean satisfiability (SAT) covers a broad spectrum of applications. Nowadays, it is an active research area finding its applications in many contexts like planning decision, cryptology, computational biology, hardware and software analysis. Hence, the development of approaches allowing to handle increasingly challenging SAT problems has become a major focus: during the past eight years, SAT solving has been the main subject of my research work. This talk presents some of the main results we obtained in the field.
<br />
<br />
</p><p><small>Souheib Baarir est Docteur en informatique de l'Université de Paris VI depuis 2007 et a obtenu son HDR à Sorbonne Université en 2019. Le thème de ses recherches s'inscrit dans le cadre des méthodes formelles de vérification des systèmes concurrents. En particulier, il s’intéresse aux méthodes permettant d’optimiser la vérification en exploitant le parallélisme et/ou les propriétés de symétries apparaissant dans de tels systèmes.</small>
<br />
<br />
<a rel="nofollow" class="external text" href="https://www.lip6.fr/actualite/personnes-fiche.php?ident=P617">https://www.lip6.fr/actualite/personnes-fiche.php?ident=P617</a>
</p></div>Mon, 15 Mar 2021 11:28:02 GMTBotNewsEntry (2021/03/05)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/03/05)
https://www.lrde.epita.fr/wiki/NewsEntry_(2021/03/05)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2021/03/05)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>Seminar on « Mathematical morphology, AI and astrometry » held at EPITA
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>E. Puybareau and G. Tochon from LRDE invite the <a rel="nofollow" class="external text" href="https://www.imcce.fr/recherche/equipes/pegase/">Pegase team from IMCCE</a> to present the respective themes of the two communities (image processing and AI for the former, astronomy for the latter) and to discuss their possible interactions.
</td></tr>
<tr>
<th>Date
</th>
<td>2021/03/05
</td></tr></tbody></table>
</div>Mon, 08 Mar 2021 11:34:02 GMTDanielaStability of the Tree of Shapes to Additive Noise
https://www.lrde.epita.fr/wiki/Publications/boutry.21.dgmm.3
https://www.lrde.epita.fr/wiki/Publications/boutry.21.dgmm.3<div class="mw-parser-output"><p><a class="mw-selflink selflink">Stability of the Tree of Shapes to Additive Noise</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, <a href="/wiki/User:Gtochon" title="User:Gtochon">Guillaume Tochon</a></dd>
<dt>Where</dt>
<dd>Proceedings of the IAPR International Conference on Discrete Geometry and Mathematical Morphology (DGMM)</dd>
<dt>Place</dt>
<dd>Uppsala, Sweden</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer&action=edit&redlink=1" class="new" title="Springer (page does not exist)">Springer</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Date</dt>
<dd>2021-03-02</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>The tree of shapes (ToS) is a famous self-dual hierarchical structure in mathematical morphology, which represents the inclusion relationship of the shapes (i.e. the interior of the level lines with holes filled) in a grayscale image. The ToS has already found numerous applications in image processing tasks, such as grain filtering, contour extraction, image simplificationand so on. Its structure consistency is bound to the cleanliness of the level lines, which are themselves deeply affected by the presence of noise within the image. However, according to our knowledge, no one has measured before how resistant to (additive) noise this hierarchical structure is. In this paper, we propose and compare several measures to evaluate the stability of the ToS structure to noise.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/boutry.21.dgmm.3.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ boutry.21.dgmm.3,
author = {Nicolas Boutry and Guillaume Tochon},
title = {Stability of the Tree of Shapes to Additive Noise},
booktitle = {Proceedings of the IAPR International Conference on
Discrete Geometry and Mathematical Morphology (DGMM)},
year = 2021,
month = may,
address = {Uppsala, Sweden},
publisher = {Springer},
series = {Lecture Notes in Computer Science},
volume = {12708},
pages = {365--377},
abstract = {The tree of shapes (ToS) is a famous self-dual
hierarchical structure in mathematical morphology, which
represents the inclusion relationship of the shapes
(\textit{i.e.} the interior of the level lines with holes
filled) in a grayscale image. The ToS has already found
numerous applications in image processing tasks, such as
grain filtering, contour extraction, image simplification,
and so on. Its structure consistency is bound to the
cleanliness of the level lines, which are themselves deeply
affected by the presence of noise within the image.
However, according to our knowledge, no one has measured
before how resistant to (additive) noise this hierarchical
structure is. In this paper, we propose and compare several
measures to evaluate the stability of the ToS structure to
noise.},
doi = {10.1007/978-3-030-76657-3_26}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Sun, 05 Dec 2021 04:10:19 GMTBotA New Matching Algorithm between Trees of Shapes and its Application to Brain Tumor Segmentation
https://www.lrde.epita.fr/wiki/Publications/boutry.21.dgmm.2
https://www.lrde.epita.fr/wiki/Publications/boutry.21.dgmm.2<div class="mw-parser-output"><p><a class="mw-selflink selflink">A New Matching Algorithm between Trees of Shapes and its Application to Brain Tumor Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a></dd>
<dt>Where</dt>
<dd>Proceedings of the IAPR International Conference on Discrete Geometry and Mathematical Morphology (DGMM)</dd>
<dt>Place</dt>
<dd>Uppsala, Sweden</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer&action=edit&redlink=1" class="new" title="Springer (page does not exist)">Springer</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Date</dt>
<dd>2021-03-02</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Many approaches exist to compute the distance between two trees in pattern recognition. These trees can be structures with or without values on their nodes or edges. Howevernone of these distances take into account the shapes possibly associated to the nodes of the tree. For this reason, we propose in this paper a new distance between two trees of shapes based on the Hausdorff distance. This distance allows us to make inexact tree matching and to compute what we call residual trees, representing where two trees differ. We will also see that thanks to these residual trees, we can obtain good results in matter of brain tumor segmentation. This segmentation does not provide only a segmentation but also the tree of shapes corresponding to the segmentation and its depth map.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/boutry.21.dgmm.2.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ boutry.21.dgmm.2,
author = {Nicolas Boutry and Thierry G\'eraud},
title = {A New Matching Algorithm between Trees of Shapes and its
Application to Brain Tumor Segmentation},
booktitle = {Proceedings of the IAPR International Conference on
Discrete Geometry and Mathematical Morphology (DGMM)},
year = 2021,
month = may,
pages = {67--78},
address = {Uppsala, Sweden},
series = {Lecture Notes in Computer Science},
volume = {12708},
publisher = {Springer},
abstract = {Many approaches exist to compute the distance between two
trees in pattern recognition. These trees can be structures
with or without values on their nodes or edges. However,
none of these distances take into account the shapes
possibly associated to the nodes of the tree. For this
reason, we propose in this paper a new distance between two
trees of shapes based on the Hausdorff distance. This
distance allows us to make inexact tree matching and to
compute what we call residual trees, representing where two
trees differ. We will also see that thanks to these
residual trees, we can obtain good results in matter of
brain tumor segmentation. This segmentation does not
provide only a segmentation but also the tree of shapes
corresponding to the segmentation and its depth map.},
doi = {10.1007/978-3-030-76657-3_4}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:54:22 GMTBotAn Equivalence Relation between Morphological Dynamics and Persistent Homology in n-D
https://www.lrde.epita.fr/wiki/Publications/boutry.21.dgmm.1
https://www.lrde.epita.fr/wiki/Publications/boutry.21.dgmm.1<div class="mw-parser-output"><p><a class="mw-selflink selflink">An Equivalence Relation between Morphological Dynamics and Persistent Homology in n-D</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd><a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a>, Laurent Najman</dd>
<dt>Where</dt>
<dd>Proceedings of the IAPR International Conference on Discrete Geometry and Mathematical Morphology (DGMM)</dd>
<dt>Place</dt>
<dd>Uppsala, Sweden</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer&action=edit&redlink=1" class="new" title="Springer (page does not exist)">Springer</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Date</dt>
<dd>2021-03-02</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>In Mathematical Morphology (MM), dynamics are used to compute markers to proceed for example to watershed-based image decomposition. At the same time, persistence is a concept coming from Persistent Homology (PH) and Morse Theory (MT) and represents the stability of the extrema of a Morse function. Since these concepts are similar on Morse functions, we studied their relationship and we found, and proved, that they are equal on 1D Morse functions. Here, we propose to extend this proof to <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a601995d55609f2d9f5e233e36fbe9ea26011b3b" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.395ex; height:1.676ex;" alt="{\displaystyle n}"/></span>-D, <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n\geq 2}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
<mo>≥<!-- ≥ --></mo>
<mn>2</mn>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n\geq 2}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/e6bf67f9d06ca3af619657f8d20ee1322da77174" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.505ex; width:5.656ex; height:2.343ex;" alt="{\displaystyle n\geq 2}"/></span>, showing that this equality can be applied to <span class="mwe-math-element"><span class="mwe-math-mathml-inline mwe-math-mathml-a11y" style="display: none;"><math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n}">
<semantics>
<mrow class="MJX-TeXAtom-ORD">
<mstyle displaystyle="true" scriptlevel="0">
<mi>n</mi>
</mstyle>
</mrow>
<annotation encoding="application/x-tex">{\displaystyle n}</annotation>
</semantics>
</math></span><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a601995d55609f2d9f5e233e36fbe9ea26011b3b" class="mwe-math-fallback-image-inline" aria-hidden="true" style="vertical-align: -0.338ex; width:1.395ex; height:1.676ex;" alt="{\displaystyle n}"/></span>-D images and not only to 1D functions. This is a step further to show how much MM and MT are related.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/boutry.21.dgmm.1.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ boutry.21.dgmm.1,
author = {Nicolas Boutry and Thierry G\'eraud and Laurent Najman},
title = {An Equivalence Relation between Morphological Dynamics and
Persistent Homology in {$n$-D}},
booktitle = {Proceedings of the IAPR International Conference on
Discrete Geometry and Mathematical Morphology (DGMM)},
year = 2021,
month = may,
address = {Uppsala, Sweden},
series = {Lecture Notes in Computer Science},
volume = {12708},
publisher = {Springer},
pages = {525--537},
abstract = {In Mathematical Morphology (MM), dynamics are used to
compute markers to proceed for example to watershed-based
image decomposition. At the same time, persistence is a
concept coming from Persistent Homology (PH) and Morse
Theory (MT) and represents the stability of the extrema of
a Morse function. Since these concepts are similar on Morse
functions, we studied their relationship and we found, and
proved, that they are equal on 1D Morse functions. Here, we
propose to extend this proof to $n$-D, $n \geq 2$, showing
that this equality can be applied to $n$-D images and not
only to 1D functions. This is a step further to show how
much MM and MT are related.},
doi = {10.1007/978-3-030-76657-3_38}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:54:17 GMTBotDeep Learning for Detection and Segmentation of Artefact and Disease Instances in Gastrointestinal Endoscopy
https://www.lrde.epita.fr/wiki/Publications/boutry.21.media
https://www.lrde.epita.fr/wiki/Publications/boutry.21.media<div class="mw-parser-output"><p><a class="mw-selflink selflink">Deep Learning for Detection and Segmentation of Artefact and Disease Instances in Gastrointestinal Endoscopy</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Sharib Ali, Mariia Dmitrieva, Noha Ghatwary, Sophia Bano, Gorkem Polat, Alptekin Temizel, Adrian Krenzer, Amar Hekalo, Yun Bo Guo, Bogdan Matuszewski, Mourad Gridach, Irina Voiculescu, Vishnusai Yoganand, Arnav Chavan, Aryan Raj, Nhan T Nguyen, Dat Q Tran, Le Duy Huynh, <a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, Shahadate Rezvy, Haijian Chen, Yoon Ho Choi, Anand Subramanian, Velmurugan Balasubramanian, Xiaohong W Gao, Hongyu Hu, Yusheng Liao, Danail Stoyanov, Christian Daul, Stefano Realdon, Renato Cannizzaro, Dominique Lamarque, Terry Tran-Nguyen, Adam Bailey, Barbara Braden, James East, Jens Rittscher</dd>
<dt>Journal</dt>
<dd>Medical Image Analysis</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Date</dt>
<dd>2021-02-24</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>The Endoscopy Computer Vision Challenge (EndoCV) is a crowd-sourcing initiative to address eminent problems in developing reliable computer aided detection and diagnosis endoscopy systems and suggest a pathway for clinical translation of technologies. Whilst endoscopy is a widely used diagnostic and treatment tool for hollow-organs, there are several core challenges often faced by endoscopistsmainly: 1) presence of multi-class artefacts that hinder their visual interpretation, and 2) difficulty in identifying subtle precancerous precursors and cancer abnormalities. Artefacts often affect the robustness of deep learning methods applied to the gastrointestinal tract organs as they can be confused with tissue of interest. EndoCV2020 challenges are designed to address research questions in these remits. In this paper, we present a summary of methods developed by the top 17 teams and provide an objective comparison of state-of-the-art methods and methods designed by the participants for two sub-challenges: i) artefact detection and segmentation (EAD2020), and ii) disease detection and segmentation (EDD2020). Multi-center, multi-organ, multi-class, and multi-modal clinical endoscopy datasets were compiled for both EAD2020 and EDD2020 sub-challenges. The out-of-sample generalization ability of detection algorithms was also evaluated. Whilst most teams focused on accuracy improvements, only a few methods hold credibility for clinical usability. The best performing teams provided solutions to tackle class imbalance, and variabilities in size, origin, modality and occurrences by exploring data augmentation, data fusion, and optimal class thresholding techniques.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/boutry.21.media.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ boutry.21.media,
author = {Sharib Ali and Mariia Dmitrieva and Noha Ghatwary and
Sophia Bano and Gorkem Polat and Alptekin Temizel and
Adrian Krenzer and Amar Hekalo and Yun Bo Guo and Bogdan
Matuszewski and Mourad Gridach and Irina Voiculescu and
Vishnusai Yoganand and Arnav Chavan and Aryan Raj and Nhan
T. Nguyen and Dat Q. Tran and Le Duy Huynh and Nicolas
Boutry and Shahadate Rezvy and Haijian Chen and Yoon Ho
Choi and Anand Subramanian and Velmurugan Balasubramanian
and Xiaohong W. Gao and Hongyu Hu and Yusheng Liao and
Danail Stoyanov and Christian Daul and Stefano Realdon and
Renato Cannizzaro and Dominique Lamarque and Terry
Tran-Nguyen and Adam Bailey and Barbara Braden and James
East and Jens Rittscher},
title = {Deep Learning for Detection and Segmentation of Artefact
and Disease Instances in Gastrointestinal Endoscopy},
journal = {Medical Image Analysis},
number = {102002},
year = {2021},
month = may,
doi = {10.1016/j.media.2021.102002},
abstract = {The Endoscopy Computer Vision Challenge (EndoCV) is a
crowd-sourcing initiative to address eminent problems in
developing reliable computer aided detection and diagnosis
endoscopy systems and suggest a pathway for clinical
translation of technologies. Whilst endoscopy is a widely
used diagnostic and treatment tool for hollow-organs, there
are several core challenges often faced by endoscopists,
mainly: 1) presence of multi-class artefacts that hinder
their visual interpretation, and 2) difficulty in
identifying subtle precancerous precursors and cancer
abnormalities. Artefacts often affect the robustness of
deep learning methods applied to the gastrointestinal tract
organs as they can be confused with tissue of interest.
EndoCV2020 challenges are designed to address research
questions in these remits. In this paper, we present a
summary of methods developed by the top 17 teams and
provide an objective comparison of state-of-the-art methods
and methods designed by the participants for two
sub-challenges: i) artefact detection and segmentation
(EAD2020), and ii) disease detection and segmentation
(EDD2020). Multi-center, multi-organ, multi-class, and
multi-modal clinical endoscopy datasets were compiled for
both EAD2020 and EDD2020 sub-challenges. The out-of-sample
generalization ability of detection algorithms was also
evaluated. Whilst most teams focused on accuracy
improvements, only a few methods hold credibility for
clinical usability. The best performing teams provided
solutions to tackle class imbalance, and variabilities in
size, origin, modality and occurrences by exploring data
augmentation, data fusion, and optimal class thresholding
techniques.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:54:30 GMTBotOn Some Associations Between Mathematical Morphology and Artificial Intelligence
https://www.lrde.epita.fr/wiki/Publications/bloch.21.dgmm
https://www.lrde.epita.fr/wiki/Publications/bloch.21.dgmm<div class="mw-parser-output"><p><a class="mw-selflink selflink">On Some Associations Between Mathematical Morphology and Artificial Intelligence</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Isabelle Bloch, Samy Blusseau, Ramón Pino Pérez, <a href="/wiki/User:Elodie" title="User:Elodie">Élodie Puybareau</a>, <a href="/wiki/User:Gtochon" title="User:Gtochon">Guillaume Tochon</a></dd>
<dt>Where</dt>
<dd>Proceedings of the IAPR International Conference on Discrete Geometry and Mathematical Morphology (DGMM)</dd>
<dt>Place</dt>
<dd>Uppsala, Sweden</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer&action=edit&redlink=1" class="new" title="Springer (page does not exist)">Springer</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-02-16</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>This paper aims at providing an overview of the use of mathematical morphology, in its algebraic setting, in several fields of artificial intelligence (AI). Three domains of AI will be covered. In the first domainmathematical morphology operators will be expressed in some logics (propositional, modal, description logics) to answer typical questions in knowledge representation and reasoning, such as revision, fusion, explanatory relationssatisfying usual postulates. In the second domain, spatial reasoning will benefit from spatial relations modeled using fuzzy sets and morphological operators, with applications in model-based image understanding. In the third domaininteractions between mathematical morphology and deep learning will be detailed. Morphological neural networks were introduced as an alternative to classical architectures, yielding a new geometry in decision surfaces. Deep networks were also trained to learn morphological operators and pipelines, and morphological algorithms were used as companion tools to machine learning, for pre/post processing or even regularization purposes. These ideas have known a large resurgence in the last few years and new ones are emerging.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ bloch.21.dgmm,
doi = {10.1007/978-3-030-76657-3_33},
author = {Isabelle Bloch and Samy Blusseau and Ram\'on {Pino
P\'erez} and \'Elodie Puybareau and Guillaume Tochon},
editor = {Lindblad, Joakim and Malmberg, Filip and Sladoje,
Nata{\v{s}}a},
title = {On Some Associations Between Mathematical Morphology and
Artificial Intelligence},
booktitle = {Proceedings of the IAPR International Conference on
Discrete Geometry and Mathematical Morphology (DGMM)},
year = {2021},
address = {Uppsala, Sweden},
series = {Lecture Notes in Computer Science},
volume = {12708},
publisher = {Springer},
pages = {457--469},
month = may,
abstract = {This paper aims at providing an overview of the use of
mathematical morphology, in its algebraic setting, in
several fields of artificial intelligence (AI). Three
domains of AI will be covered. In the first domain,
mathematical morphology operators will be expressed in some
logics (propositional, modal, description logics) to answer
typical questions in knowledge representation and
reasoning, such as revision, fusion, explanatory relations,
satisfying usual postulates. In the second domain, spatial
reasoning will benefit from spatial relations modeled using
fuzzy sets and morphological operators, with applications
in model-based image understanding. In the third domain,
interactions between mathematical morphology and deep
learning will be detailed. Morphological neural networks
were introduced as an alternative to classical
architectures, yielding a new geometry in decision
surfaces. Deep networks were also trained to learn
morphological operators and pipelines, and morphological
algorithms were used as companion tools to machine
learning, for pre/post processing or even regularization
purposes. These ideas have known a large resurgence in the
last few years and new ones are emerging.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:53:58 GMTBotCombining Deep Learning and Mathematical Morphology for Historical Map Segmentation
https://www.lrde.epita.fr/wiki/Publications/chen.21.dgmm
https://www.lrde.epita.fr/wiki/Publications/chen.21.dgmm<div class="mw-parser-output"><p><a class="mw-selflink selflink">Combining Deep Learning and Mathematical Morphology for Historical Map Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Yizi Chen, <a href="/wiki/User:Carlinet" title="User:Carlinet">Edwin Carlinet</a>, <a href="/wiki/User:Chazalon" title="User:Chazalon">Joseph Chazalon</a>, Clément Mallet, Bertrand Duménieu, Julien Perret</dd>
<dt>Where</dt>
<dd>Proceedings of the IAPR International Conference on Discrete Geometry and Mathematical Morphology (DGMM)</dd>
<dt>Place</dt>
<dd>Uppsala, Sweden</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer&action=edit&redlink=1" class="new" title="Springer (page does not exist)">Springer</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-02-16</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>The digitization of historical maps enables the study of ancient, fragile, unique, and hardly accessible information sources. Main map features can be retrieved and tracked through the time for subsequent thematic analysis. The goal of this work is the vectorization step, i.e., the extraction of vector shapes of the objects of interest from raster images of maps. We are particularly interested in closed shape detection such as buildings, building blocksgardens, rivers, etc. in order to monitor their temporal evolution. Historical map images present significant pattern recognition challenges. The extraction of closed shapes by using traditional Mathematical Morphology (MM) is highly challenging due to the overlapping of multiple map features and texts. Moreover, state-of-the-art Convolutional Neural Networks (CNN) are perfectly designed for content image filtering but provide no guarantee about closed shape detection. Also, the lack of textural and color information of historical maps makes it hard for CNN to detect shapes that are represented by only their boundaries. Our contribution is a pipeline that combines the strengths of CNN (efficient edge detection and filtering) and MM (guaranteed extraction of closed shapes) in order to achieve such a task. The evaluation of our approach on a public dataset shows its effectiveness for extracting the closed boundaries of objects in historical maps.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/chen.2021.dgmm.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ chen.21.dgmm,
author = {Yizi Chen and Edwin Carlinet and Joseph Chazalon and
Cl\'ement Mallet and Bertrand Dum\'enieu and Julien Perret},
title = {Combining Deep Learning and Mathematical Morphology for
Historical Map Segmentation},
booktitle = {Proceedings of the IAPR International Conference on
Discrete Geometry and Mathematical Morphology (DGMM)},
year = {2021},
series = {Lecture Notes in Computer Science},
volume = {12708},
month = may,
address = {Uppsala, Sweden},
publisher = {Springer},
pages = {79--92},
abstract = {The digitization of historical maps enables the study of
ancient, fragile, unique, and hardly accessible information
sources. Main map features can be retrieved and tracked
through the time for subsequent thematic analysis. The goal
of this work is the vectorization step, i.e., the
extraction of vector shapes of the objects of interest from
raster images of maps. We are particularly interested in
closed shape detection such as buildings, building blocks,
gardens, rivers, etc. in order to monitor their temporal
evolution. Historical map images present significant
pattern recognition challenges. The extraction of closed
shapes by using traditional Mathematical Morphology (MM) is
highly challenging due to the overlapping of multiple map
features and texts. Moreover, state-of-the-art
Convolutional Neural Networks (CNN) are perfectly designed
for content image filtering but provide no guarantee about
closed shape detection. Also, the lack of textural and
color information of historical maps makes it hard for CNN
to detect shapes that are represented by only their
boundaries. Our contribution is a pipeline that combines
the strengths of CNN (efficient edge detection and
filtering) and MM (guaranteed extraction of closed shapes)
in order to achieve such a task. The evaluation of our
approach on a public dataset shows its effectiveness for
extracting the closed boundaries of objects in historical
maps.},
note = {Accepted},
doi = {10.1007/978-3-030-76657-3_5}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:55:07 GMTBotGoing beyond p-convolutions to learn grayscale morphological operators
https://www.lrde.epita.fr/wiki/Publications/kirszenberg.21.dgmm
https://www.lrde.epita.fr/wiki/Publications/kirszenberg.21.dgmm<div class="mw-parser-output"><p><a class="mw-selflink selflink">Going beyond p-convolutions to learn grayscale morphological operators</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Alexandre Kirszenberg, <a href="/wiki/User:Gtochon" title="User:Gtochon">Guillaume Tochon</a>, <a href="/wiki/User:Elodie" title="User:Elodie">Élodie Puybareau</a>, Jesus Angulo</dd>
<dt>Where</dt>
<dd>Proceedings of the IAPR International Conference on Discrete Geometry and Mathematical Morphology (DGMM)</dd>
<dt>Place</dt>
<dd>Uppsala, Sweden</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=Springer&action=edit&redlink=1" class="new" title="Springer (page does not exist)">Springer</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2021-02-16</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Integrating mathematical morphology operations within deep neural networks has been subject to increasing attention lately. However, replacing standard convolution layers with erosions or dilations is particularly challenging because the min and max operations are not differentiable. Relying on the asymptotic behavior of the counter-harmonic meanp-convolutional layers were proposed as a possible workaround to this issue since they can perform pseudo-dilation or pseudo-erosion operations (depending on the value of their inner parameter p), and very promising results were reported. In this work, we present two new morphological layers based on the same principle as the p-convolutional layer while circumventing its principal drawbacks, and demonstrate their potential interest in further implementations within deep convolutional neural network architectures.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/kirszie.2021.dgmm.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ kirszenberg.21.dgmm,
author = {Alexandre Kirszenberg and Guillaume Tochon and \'{E}lodie
Puybareau and Jesus Angulo},
title = {Going beyond p-convolutions to learn grayscale
morphological operators},
booktitle = {Proceedings of the IAPR International Conference on
Discrete Geometry and Mathematical Morphology (DGMM)},
year = {2021},
series = {Lecture Notes in Computer Science},
volume = {12708},
month = may,
address = {Uppsala, Sweden},
publisher = {Springer},
pages = {470--482},
abstract = {Integrating mathematical morphology operations within deep
neural networks has been subject to increasing attention
lately. However, replacing standard convolution layers with
erosions or dilations is particularly challenging because
the min and max operations are not differentiable. Relying
on the asymptotic behavior of the counter-harmonic mean,
p-convolutional layers were proposed as a possible
workaround to this issue since they can perform
pseudo-dilation or pseudo-erosion operations (depending on
the value of their inner parameter p), and very promising
results were reported. In this work, we present two new
morphological layers based on the same principle as the
p-convolutional layer while circumventing its principal
drawbacks, and demonstrate their potential interest in
further implementations within deep convolutional neural
network architectures.},
doi = {10.1007/978-3-030-76657-3_34}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:56:52 GMTBotSeminar/2021-02-10
https://www.lrde.epita.fr/wiki/Seminar/2021-02-10
https://www.lrde.epita.fr/wiki/Seminar/2021-02-10<div class="mw-parser-output"><p><a class="mw-selflink selflink">Seminar/2021-02-10</a>
</p></div><div class="mw-parser-output"><h3><span id="Mercredi_10_février_2021,_11h_-_12h,_{\small_https://meet.jit.si/Seminaire$_$LRDE$_$Uli"></span><span class="mw-headline" id="Mercredi_10_f.C3.A9vrier_2021.2C_11h_-_12h.2C_.7B.5Csmall_https:.2F.2Fmeet.jit.si.2FSeminaire.24_.24LRDE.24_.24Uli"><a class="mw-selflink selflink"> Mercredi 10 février 2021, 11h - 12h, {\small https://meet.jit.si/Seminaire$_$LRDE$_$Uli</a></span></h3>
<p><br />
</p>
<h4><span class="mw-headline" id="Generating_Posets_Beyond_N">Generating Posets Beyond N</span></h4>
<p><i>Uli Fahrenberg, Ecole Polytechnique</i>
<br />
<br />
</p><p>We introduce iposets - posets with interfaces - equipped with a novel gluing
composition along interfaces and the standard parallel composition. We study
their basic algebraic properties as well as the hierarchy of gluing-parallel
posets generated from singletons by finitary applications of the two
compositions. We show that not only series-parallel posets, but also
interval orders, which seem more interesting for modeling concurrent and
distributed systems, can be generated, but not all posets. Generating posets
is also important for constructing free algebras for concurrent semi-rings
and Kleene algebras that allow compositional reasoning about such systems.
<br />
<br />
</p><p><small>Ulrich (Uli) Fahrenberg holds a PhD in mathematics from Aalborg University, Denmark. He has started his career in computer science as an assistant professor at Aalborg University. Afterwards he has worked as a postdoc at Inria Rennes, France, and since 2016 he is a researcher at the computer science lab at École polytechnique in Palaiseau, France. Uli Fahrenberg works in algebraic topology, concurrency theory, real-time verification, and general quantitative verification. He has published more than 80 papers in computer science and mathematics. He has been a member of numerous program committees, and since 2016 he is a reviewer for AMS Mathematical Reviews.</small>
<br />
<br />
<a rel="nofollow" class="external text" href="http://www.lix.polytechnique.fr/~uli/bio.html">http://www.lix.polytechnique.fr/~uli/bio.html</a>
</p></div>Tue, 26 Jan 2021 19:07:08 GMTBotSeminar/2020-12-16
https://www.lrde.epita.fr/wiki/Seminar/2020-12-16
https://www.lrde.epita.fr/wiki/Seminar/2020-12-16<div class="mw-parser-output"><p><a class="mw-selflink selflink">Seminar/2020-12-16</a>
</p></div><div class="mw-parser-output"><h3><span id="Mercredi_16_décembre_2020,_11h_-_12h,_{\small_https://eu.bbcollab.com/collab/ui/session/guest/95a72a9dc7b0405c8c281ea3157e9637}"></span><span class="mw-headline" id="Mercredi_16_d.C3.A9cembre_2020.2C_11h_-_12h.2C_.7B.5Csmall_https:.2F.2Feu.bbcollab.com.2Fcollab.2Fui.2Fsession.2Fguest.2F95a72a9dc7b0405c8c281ea3157e9637.7D"><a class="mw-selflink selflink"> Mercredi 16 décembre 2020, 11h - 12h, {\small https://eu.bbcollab.com/collab/ui/session/guest/95a72a9dc7b0405c8c281ea3157e9637}</a></span></h3>
<p><br />
</p>
<h4><span class="mw-headline" id="Diagnosis_and_Opacity_in_Partially_Observable_Systems">Diagnosis and Opacity in Partially Observable Systems</span></h4>
<p><i>Stefan Schwoon, ENS Paris-Saclay</i>
<br />
<br />
</p><p>In a partially observable system, diagnosis is the task of detecting certain events, for instance fault occurrences. In the presence of hostile observers, on the other hand, one is interested in rendering a system opaque, i.e. making it impossible to detect certain "secret" events. The talk will present some decidability and complexity results for these two problems
when the system is represented as a finite automaton or a Petri net. We then also consider the problem of active diagnosis, where the observer has some control over the system. In this context, we study problems such as the computational complexity of the synthesis problem, the memory required for the controller, and the delay between a fault occurrence and its detection by the diagnoser. The talk is based on joint work with B. Bérard, S. Haar, S. Haddad, T. Melliti, and S. Schmitz.
<br />
<br />
</p><p><small>Stefan Schwoon studied Computer Science at the University of Hildesheim and received a PhD from the Technical University of Munich in 2002. He held the position of Scientific Assistent at the University of Stuttgart from 2002 to 2007, and at the Technical University in Munich from 2007 to 2009. He is currently Associate Professor (Maître de conférences) at Laboratoire Spécification et Vérification (LSV), ENS Paris-Saclay, and a member of the INRIA team Mexico. His research interests include model checking and diagnosis on concurrent and partially-observable systems.</small>
<br />
<br />
<a rel="nofollow" class="external text" href="http://www.lsv.fr/~schwoon/">http://www.lsv.fr/~schwoon/</a>
</p></div>Tue, 26 Jan 2021 19:07:07 GMTBotNewsEntry (2020/11/16)
https://www.lrde.epita.fr/wiki/NewsEntry_(2020/11/16)
https://www.lrde.epita.fr/wiki/NewsEntry_(2020/11/16)<div class="mw-parser-output"><p><a class="mw-selflink selflink">NewsEntry (2020/11/16)</a>
</p></div><div class="mw-parser-output"><table class="wikitable">
<tbody><tr>
<th>Title
</th>
<td>The LRDE hosts a new member, Baptiste Esteban, who joins the <a href="/wiki/Olena" title="Olena">Olena</a> team for his PhD studies.
</td></tr>
<tr>
<th>Sub-Title
</th>
<td>After completing <a rel="nofollow" class="external text" href="https://www.epita.fr/nos-formations/diplome-ingenieur/cycle-ingenieur/les-majeures/">EPITA's IMAGE and RDI double major</a>, Baptiste is back at LRDE for his PhD. Having worked on noise estimation in natural images with mathematical morphology approaches, he will now focus on how to conciliate genericity and performance of image processing algorithms in dynamic contexts, especially noise estimation as a validation framework.
</td></tr>
<tr>
<th>Date
</th>
<td>2020/11/16
</td></tr></tbody></table>
</div>Wed, 18 Nov 2020 09:45:01 GMTDanielaA Global Benchmark of Algorithms for Segmenting the Left Atrium from Late Gadolinium-Enhanced Cardiac Magnetic Resonance Imaging
https://www.lrde.epita.fr/wiki/Publications/xiong.20.media
https://www.lrde.epita.fr/wiki/Publications/xiong.20.media<div class="mw-parser-output"><p><a class="mw-selflink selflink">A Global Benchmark of Algorithms for Segmenting the Left Atrium from Late Gadolinium-Enhanced Cardiac Magnetic Resonance Imaging</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Zhaohan Xiong, Qing Xia, Zhiqiang Hu, Ning Huang, Cheng Bian, Yefeng Zheng, Sulaiman Vesal, Nishant Ravikumar, Andreas Maier, Xin Yang, Pheng-Ann Heng, Dong Ni, Caizi Li, Qianqian Tong, Weixin Si, <a href="/wiki/User:Elodie" title="User:Elodie">Élodie Puybareau</a>, Younes Khoudli, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a>, Chen Chen, Wenjia Bai, Daniel Rueckert, Lingchao Xu, Xiahai Zhuang, Xinzhe Luo, Shuman Jia, Maxime Sermesant, Yashu Liu, Kuanquan Wang, Davide Borra, Alessandro Masci, Cristiana Corsi, Coen de Vente, Mitko Veta, Rashed Karim, Chandrakanth Jayachandran Preetha, Sandy Engelhardt, Menyun Qiao, Yuanyuan Wang, Qian Tao, Marta Nunez-Garcia, Oscar Camara, Nicolo Savioli, Pablo Lamata, Jichao Zhao</dd>
<dt>Journal</dt>
<dd>Medical Image Analysis</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Date</dt>
<dd>2020-11-10</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Segmentation of medical images, particularly late gadolinium-enhanced magnetic resonance imaging (LGE-MRI) used for visualizing diseased atrial structures, is a crucial first step for ablation treatment of atrial fibrillation. However, direct segmentation of LGE-MRIs is challenging due to the varying intensities caused by contrast agents. Since most clinical studies have relied on manual, labor-intensive approaches, automatic methods are of high interest, particularly optimized machine learning approaches. To address this, we organized the 2018 Left Atrium Segmentation Challenge using 154 3D LGE-MRIscurrently the world's largest atrial LGE-MRI dataset, and associated labels of the left atrium segmented by three medical experts, ultimately attracting the participation of 27 international teams. In this paper, extensive analysis of the submitted algorithms using technical and biological metrics was performed by undergoing subgroup analysis and conducting hyper-parameter analysis, offering an overall picture of the major design choices of convolutional neural networks (CNNs) and practical considerations for achieving state-of-the-art left atrium segmentation. Results show that the top method achieved a Dice score of 93.2% and a mean surface to surface distance of 0.7 mm, significantly outperforming prior state-of-the-art. Particularly, our analysis demonstrated that double sequentially used CNNsin which a first CNN is used for automatic region-of-interest localization and a subsequent CNN is used for refined regional segmentation, achieved superior results than traditional methods and machine learning approaches containing single CNNs. This large-scale benchmarking study makes a significant step towards much-improved segmentation methods for atrial LGE-MRIs, and will serve as an important benchmark for evaluating and comparing the future works in the field. Furthermore, the findings from this study can potentially be extended to other imaging datasets and modalities, having an impact on the wider medical imaging community.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ xiong.20.media,
title = {A Global Benchmark of Algorithms for Segmenting the Left
Atrium from Late Gadolinium-Enhanced Cardiac Magnetic
Resonance Imaging},
journal = {Medical Image Analysis},
volume = {67},
pages = {101832},
year = {2021},
month = jan,
issn = {1361-8415},
doi = {10.1016/j.media.2020.101832},
author = {Zhaohan Xiong and Qing Xia and Zhiqiang Hu and Ning Huang
and Cheng Bian and Yefeng Zheng and Sulaiman Vesal and
Nishant Ravikumar and Andreas Maier and Xin Yang and
Pheng-Ann Heng and Dong Ni and Caizi Li and Qianqian Tong
and Weixin Si and \'Elodie Puybareau and Younes Khoudli and
Thierry G\'{e}raud and Chen Chen and Wenjia Bai and Daniel
Rueckert and Lingchao Xu and Xiahai Zhuang and Xinzhe Luo
and Shuman Jia and Maxime Sermesant and Yashu Liu and
Kuanquan Wang and Davide Borra and Alessandro Masci and
Cristiana Corsi and Coen {de Vente} and Mitko Veta and
Rashed Karim and Chandrakanth Jayachandran Preetha and
Sandy Engelhardt and Menyun Qiao and Yuanyuan Wang and Qian
Tao and Marta Nunez-Garcia and Oscar Camara and Nicolo
Savioli and Pablo Lamata and Jichao Zhao},
keywords = {Left atrium, Convolutional neural networks, Late
gadolinium-enhanced magnetic resonance imaging, Image
segmentation},
abstract = {Segmentation of medical images, particularly late
gadolinium-enhanced magnetic resonance imaging (LGE-MRI)
used for visualizing diseased atrial structures, is a
crucial first step for ablation treatment of atrial
fibrillation. However, direct segmentation of LGE-MRIs is
challenging due to the varying intensities caused by
contrast agents. Since most clinical studies have relied on
manual, labor-intensive approaches, automatic methods are
of high interest, particularly optimized machine learning
approaches. To address this, we organized the 2018 Left
Atrium Segmentation Challenge using 154 3D LGE-MRIs,
currently the world's largest atrial LGE-MRI dataset, and
associated labels of the left atrium segmented by three
medical experts, ultimately attracting the participation of
27 international teams. In this paper, extensive analysis
of the submitted algorithms using technical and biological
metrics was performed by undergoing subgroup analysis and
conducting hyper-parameter analysis, offering an overall
picture of the major design choices of convolutional neural
networks (CNNs) and practical considerations for achieving
state-of-the-art left atrium segmentation. Results show
that the top method achieved a Dice score of 93.2\% and a
mean surface to surface distance of 0.7 mm, significantly
outperforming prior state-of-the-art. Particularly, our
analysis demonstrated that double sequentially used CNNs,
in which a first CNN is used for automatic
region-of-interest localization and a subsequent CNN is
used for refined regional segmentation, achieved superior
results than traditional methods and machine learning
approaches containing single CNNs. This large-scale
benchmarking study makes a significant step towards
much-improved segmentation methods for atrial LGE-MRIs, and
will serve as an important benchmark for evaluating and
comparing the future works in the field. Furthermore, the
findings from this study can potentially be extended to
other imaging datasets and modalities, having an impact on
the wider medical imaging community.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 22 Jul 2021 09:47:23 GMTBotPAIP 2019: Liver Cancer Segmentation Challenge
https://www.lrde.epita.fr/wiki/Publications/kim.20.media
https://www.lrde.epita.fr/wiki/Publications/kim.20.media<div class="mw-parser-output"><p><a class="mw-selflink selflink">PAIP 2019: Liver Cancer Segmentation Challenge</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Yoo Jung Kim, Hyungjoon Jang, Kyoungbun Lee, Seongkeun Park, Sung-Gyu Min, Choyeon Hong, Jeong Hwan Park, Kanggeun Lee, Jisoo Kim, Wonjae Hong, Hyun Jung, Yanling Liu, Haran Rajkumar, Mahendra Khened, Ganapathy Krishnamurthi, Sen Yang, Xiyue Wang, Chang Hee Han, Jin Tae Kwak, Jianqiang Ma, Zhe Tang, Bahram Marami, Jack Zeineh, Zixu Zhao, Pheng-Ann Heng, Rudiger Schmitz, Frederic Madesta, Thomas Rosch, Rene Werner, Jie Tian, <a href="/wiki/User:Elodie" title="User:Elodie">Élodie Puybareau</a>, Matteo Bovio, Xiufeng Zhang, Yifeng Zhu, Se Young Chun, Won-Ki Jeong, Peom Park, Jinwook Choi</dd>
<dt>Journal</dt>
<dd>Medical Image Analysis</dd>
<dt>Type</dt>
<dd>article</dd>
<dt>Date</dt>
<dd>2020-11-10</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Pathology Artificial Intelligence Platform (PAIP) is a free research platform in support of pathological artificial intelligence (AI). The main goal of the platform is to construct a high-quality pathology learning data set that will allow greater accessibility. The PAIP Liver Cancer Segmentation Challenge, organized in conjunction with the Medical Image Computing and Computer Assisted Intervention Society (MICCAI 2019), is the first image analysis challenge to apply PAIP datasets. The goal of the challenge was to evaluate new and existing algorithms for automated detection of liver cancer in whole-slide images (WSIs). Additionally, the PAIP of this year attempted to address potential future problems of AI applicability in clinical settings. In the challenge, participants were asked to use analytical data and statistical metrics to evaluate the performance of automated algorithms in two different tasks. The participants were given the two different tasks: Task 1 involved investigating Liver Cancer Segmentation and Task 2 involved investigating Viable Tumor Burden Estimation. There was a strong correlation between high performance of teams on both tasks, in which teams that performed well on Task 1 also performed well on Task 2. After evaluation, we summarized the top 11 team's algorithms. We then gave pathological implications on the easily predicted images for cancer segmentation and the challenging images for viable tumor burden estimation. Out of the 231 participants of the PAIP challenge datasets, a total of 64 were submitted from 28 team participants. The submitted algorithms predicted the automatic segmentation on the liver cancer with WSIs to an accuracy of a score estimation of 0.78. The PAIP challenge was created in an effort to combat the lack of research that has been done to address Liver cancer using digital pathology. It remains unclear of how the applicability of AI algorithms created during the challenge can affect clinical diagnoses. However, the results of this dataset and evaluation metric provided has the potential to aid the development and benchmarking of cancer diagnosis and segmentation.
</p><p><br />
</p>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@Article{ kim.20.media,
title = {{PAIP} 2019: {L}iver Cancer Segmentation Challenge},
journal = {Medical Image Analysis},
volume = {67},
pages = {101854},
year = {2021},
month = jan,
issn = {1361-8415},
doi = {10.1016/j.media.2020.101854},
author = {Yoo Jung Kim and Hyungjoon Jang and Kyoungbun Lee and
Seongkeun Park and Sung-Gyu Min and Choyeon Hong and Jeong
Hwan Park and Kanggeun Lee and Jisoo Kim and Wonjae Hong
and Hyun Jung and Yanling Liu and Haran Rajkumar and
Mahendra Khened and Ganapathy Krishnamurthi and Sen Yang
and Xiyue Wang and Chang Hee Han and Jin Tae Kwak and
Jianqiang Ma and Zhe Tang and Bahram Marami and Jack Zeineh
and Zixu Zhao and Pheng-Ann Heng and Rudiger Schmitz and
Frederic Madesta and Thomas Rosch and Rene Werner and Jie
Tian and \'Elodie Puybareau and Matteo Bovio and Xiufeng
Zhang and Yifeng Zhu and Se Young Chun and Won-Ki Jeong and
Peom Park and Jinwook Choi},
keywords = {Liver cancer, Tumor burden, Digital pathology, Challenge,
Segmentation},
abstract = {Pathology Artificial Intelligence Platform (PAIP) is a
free research platform in support of pathological
artificial intelligence (AI). The main goal of the platform
is to construct a high-quality pathology learning data set
that will allow greater accessibility. The PAIP Liver
Cancer Segmentation Challenge, organized in conjunction
with the Medical Image Computing and Computer Assisted
Intervention Society (MICCAI 2019), is the first image
analysis challenge to apply PAIP datasets. The goal of the
challenge was to evaluate new and existing algorithms for
automated detection of liver cancer in whole-slide images
(WSIs). Additionally, the PAIP of this year attempted to
address potential future problems of AI applicability in
clinical settings. In the challenge, participants were
asked to use analytical data and statistical metrics to
evaluate the performance of automated algorithms in two
different tasks. The participants were given the two
different tasks: Task 1 involved investigating Liver Cancer
Segmentation and Task 2 involved investigating Viable Tumor
Burden Estimation. There was a strong correlation between
high performance of teams on both tasks, in which teams
that performed well on Task 1 also performed well on Task
2. After evaluation, we summarized the top 11 team's
algorithms. We then gave pathological implications on the
easily predicted images for cancer segmentation and the
challenging images for viable tumor burden estimation. Out
of the 231 participants of the PAIP challenge datasets, a
total of 64 were submitted from 28 team participants. The
submitted algorithms predicted the automatic segmentation
on the liver cancer with WSIs to an accuracy of a score
estimation of 0.78. The PAIP challenge was created in an
effort to combat the lack of research that has been done to
address Liver cancer using digital pathology. It remains
unclear of how the applicability of AI algorithms created
during the challenge can affect clinical diagnoses.
However, the results of this dataset and evaluation metric
provided has the potential to aid the development and
benchmarking of cancer diagnosis and segmentation.}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Thu, 22 Jul 2021 09:46:13 GMTBotDo not Treat Boundaries and Regions Differently: An Example on Heart Left Atrial Segmentation
https://www.lrde.epita.fr/wiki/Publications/zhao.20.icpr.2
https://www.lrde.epita.fr/wiki/Publications/zhao.20.icpr.2<div class="mw-parser-output"><p><a class="mw-selflink selflink">Do not Treat Boundaries and Regions Differently: An Example on Heart Left Atrial Segmentation</a>
</p></div><div class="mw-parser-output"><div class="sideBox">
<dl><dt>Authors</dt>
<dd>Zhou Zhao, <a href="/wiki/User:Nboutry" title="User:Nboutry">Nicolas Boutry</a>, <a href="/wiki/User:Elodie" title="User:Elodie">Élodie Puybareau</a>, <a href="/wiki/User:Theo" title="User:Theo">Thierry Géraud</a></dd>
<dt>Where</dt>
<dd>Proceedings of the 25th International Conference on Pattern Recognition (ICPR)</dd>
<dt>Place</dt>
<dd>Milan, Italy</dd>
<dt>Type</dt>
<dd>inproceedings</dd>
<dt>Publisher</dt>
<dd><a href="/index.php?title=IEEE&action=edit&redlink=1" class="new" title="IEEE (page does not exist)">IEEE</a></dd>
<dt>Projects</dt>
<dd><a href="/wiki/Olena" title="Olena">Olena</a></dd>
<dt>Keywords</dt>
<dd>Image</dd>
<dt>Date</dt>
<dd>2020-11-02</dd></dl>
</div>
<h2><span class="mw-headline" id="Abstract">Abstract</span></h2>
<p>Atrial fibrillation is the most common heart rhythm disease. Due to a lack of understanding in matter of underlying atrial structures, current treatments are still not satisfying. Recently, with the popularity of deep learning, many segmentation methods based on fully convolutional networks have been proposed to analyze atrial structures, especially from late gadolinium-enhanced magnetic resonance imaging. However, two problems still occur: 1) segmentation results include the atrial- like background; 2) boundaries are very hard to segment. Most segmentation approaches design a specific network that mainly focuses on the regions, to the detriment of the boundaries. Therefore, this paper proposes an attention full convolutional network framework based on the ResNet-101 architecture, which focuses on boundaries as much as on regions. The additional attention module is added to have the network pay more attention on regions and then to reduce the impact of the misleading similarity of neighboring tissues. We also use a hybrid loss composed of a region loss and a boundary loss to treat boundaries and regions at the same time. We demonstrate the efficiency of the proposed approach on the MICCAI 2018 Atrial Segmentation Challenge public dataset.
</p>
<h2><span class="mw-headline" id="Documents">Documents</span></h2>
<ul><li><a rel="nofollow" class="external text" href="http://www.lrde.epita.fr/dload/papers/zhao.20.icpr.2.pdf">Paper</a></li></ul>
<h2><span id="Bibtex_(lrde.bib)"></span><span class="mw-headline" id="Bibtex_.28lrde.bib.29">Bibtex (<a rel="nofollow" class="external text" href="https://www.lrde.epita.fr/dload/papers/lrde.bib">lrde.bib</a>)</span></h2>
<p><small>
</small></p><small><pre>@InProceedings{ zhao.20.icpr.2,
author = {Zhou Zhao and Nicolas Boutry and \'Elodie Puybareau and
Thierry G\'eraud},
title = {Do not Treat Boundaries and Regions Differently: {A}n
Example on Heart Left Atrial Segmentation},
booktitle = {Proceedings of the 25th International Conference on
Pattern Recognition (ICPR)},
year = 2021,
pages = {7447--7453},
month = jan,
address = {Milan, Italy},
publisher = {IEEE},
abstract = {Atrial fibrillation is the most common heart rhythm
disease. Due to a lack of understanding in matter of
underlying atrial structures, current treatments are still
not satisfying. Recently, with the popularity of deep
learning, many segmentation methods based on fully
convolutional networks have been proposed to analyze atrial
structures, especially from late gadolinium-enhanced
magnetic resonance imaging. However, two problems still
occur: 1) segmentation results include the atrial- like
background; 2) boundaries are very hard to segment. Most
segmentation approaches design a specific network that
mainly focuses on the regions, to the detriment of the
boundaries. Therefore, this paper proposes an attention
full convolutional network framework based on the
ResNet-101 architecture, which focuses on boundaries as
much as on regions. The additional attention module is
added to have the network pay more attention on regions and
then to reduce the impact of the misleading similarity of
neighboring tissues. We also use a hybrid loss composed of
a region loss and a boundary loss to treat boundaries and
regions at the same time. We demonstrate the efficiency of
the proposed approach on the MICCAI 2018 Atrial
Segmentation Challenge public dataset.},
doi = {10.1109/ICPR48806.2021.9412755}
}</pre></small><small></small><p><small></small>
</p><p><br />
</p><p><br />
</p><p><br />
</p><p><br />
</p></div>Wed, 08 Sep 2021 08:59:09 GMTBot