Difference between revisions of "Publications/royer.17.icdar"
From LRDE
Line 5: | Line 5: | ||
| authors = E Royer, J Chazalon, M Rusiñol, F Bouchara |
| authors = E Royer, J Chazalon, M Rusiñol, F Bouchara |
||
| booktitle = Proceedings of the 14th International Conference on Document Analysis and Recognition (ICDAR) |
| booktitle = Proceedings of the 14th International Conference on Document Analysis and Recognition (ICDAR) |
||
+ | | pages = 343 to 348 |
||
+ | | address = Kyoto, Japan |
||
| abstract = Reducing the amount of keypoints used to index an image is particularly interesting to control processing time and memory usage in real-time document image matching applications, like augmented documents or smartphone applications. This paper benchmarks two keypoint selection methods on a task consisting of reducing keypoint sets extracted from document images, while preserving detection and segmentation accuracy. We first study the different forms of keypoint filtering, and we introduce the use of the CORE selection method on keypoints extracted from document images. Then, we extend a previously published benchmark by including evaluations of the new method, by adding the SURF-BRISK detection/description scheme, and by reporting processing speeds. Evaluations are conducted on the publicly available dataset of ICDAR2015 SmartDOC challenge 1. Finally, we prove that reducing the original keypoint set is always feasible and can be beneficial not only to processing speed but also to accuracy. |
| abstract = Reducing the amount of keypoints used to index an image is particularly interesting to control processing time and memory usage in real-time document image matching applications, like augmented documents or smartphone applications. This paper benchmarks two keypoint selection methods on a task consisting of reducing keypoint sets extracted from document images, while preserving detection and segmentation accuracy. We first study the different forms of keypoint filtering, and we introduce the use of the CORE selection method on keypoints extracted from document images. Then, we extend a previously published benchmark by including evaluations of the new method, by adding the SURF-BRISK detection/description scheme, and by reporting processing speeds. Evaluations are conducted on the publicly available dataset of ICDAR2015 SmartDOC challenge 1. Finally, we prove that reducing the original keypoint set is always feasible and can be beneficial not only to processing speed but also to accuracy. |
||
| lrdepaper = http://www.lrde.epita.fr/dload/papers/royer.17.icdar.pdf |
| lrdepaper = http://www.lrde.epita.fr/dload/papers/royer.17.icdar.pdf |
||
Line 21: | Line 23: | ||
Document Analysis and Recognition (ICDAR)<nowiki>}</nowiki>, |
Document Analysis and Recognition (ICDAR)<nowiki>}</nowiki>, |
||
year = <nowiki>{</nowiki>2017<nowiki>}</nowiki>, |
year = <nowiki>{</nowiki>2017<nowiki>}</nowiki>, |
||
+ | month = nov, |
||
+ | pages = <nowiki>{</nowiki>343--348<nowiki>}</nowiki>, |
||
+ | address = <nowiki>{</nowiki>Kyoto, Japan<nowiki>}</nowiki>, |
||
abstract = <nowiki>{</nowiki>Reducing the amount of keypoints used to index an image is |
abstract = <nowiki>{</nowiki>Reducing the amount of keypoints used to index an image is |
||
particularly interesting to control processing time and |
particularly interesting to control processing time and |
Revision as of 15:17, 29 October 2018
- Authors
- E Royer, J Chazalon, M Rusiñol, F Bouchara
- Where
- Proceedings of the 14th International Conference on Document Analysis and Recognition (ICDAR)
- Place
- Kyoto, Japan
- Type
- inproceedings
- Projects
- Olena
- Keywords
- Image
- Date
- 2017-07-04
Abstract
Reducing the amount of keypoints used to index an image is particularly interesting to control processing time and memory usage in real-time document image matching applications, like augmented documents or smartphone applications. This paper benchmarks two keypoint selection methods on a task consisting of reducing keypoint sets extracted from document images, while preserving detection and segmentation accuracy. We first study the different forms of keypoint filtering, and we introduce the use of the CORE selection method on keypoints extracted from document images. Then, we extend a previously published benchmark by including evaluations of the new method, by adding the SURF-BRISK detection/description scheme, and by reporting processing speeds. Evaluations are conducted on the publicly available dataset of ICDAR2015 SmartDOC challenge 1. Finally, we prove that reducing the original keypoint set is always feasible and can be beneficial not only to processing speed but also to accuracy.
Documents
Bibtex (lrde.bib)
@InProceedings{ royer.17.icdar, title = {Benchmarking Keypoint Filtering Approaches for Document Image Matching}, author = {E. Royer and J. Chazalon and M. Rusi{\~n}ol and F. Bouchara}, booktitle = {Proceedings of the 14th International Conference on Document Analysis and Recognition (ICDAR)}, year = {2017}, month = nov, pages = {343--348}, address = {Kyoto, Japan}, abstract = {Reducing the amount of keypoints used to index an image is particularly interesting to control processing time and memory usage in real-time document image matching applications, like augmented documents or smartphone applications. This paper benchmarks two keypoint selection methods on a task consisting of reducing keypoint sets extracted from document images, while preserving detection and segmentation accuracy. We first study the different forms of keypoint filtering, and we introduce the use of the CORE selection method on keypoints extracted from document images. Then, we extend a previously published benchmark by including evaluations of the new method, by adding the SURF-BRISK detection/description scheme, and by reporting processing speeds. Evaluations are conducted on the publicly available dataset of ICDAR2015 SmartDOC challenge 1. Finally, we prove that reducing the original keypoint set is always feasible and can be beneficial not only to processing speed but also to accuracy.} }