Jobs/M2 JF 2014 inpainting

From LRDE

Revision as of 11:58, 19 February 2014 by Daniela (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Inpainting in videos
Reference id

M2 JF 2014 inpainting

Dates

6 months in 2014 - starting as soon as possible

Research field

Image Processing

Related project

Olena

Advisor

Jonathan Fabrizio

General presentation of the field

Automatic text localization in natural images (still or moving) is an essential task in many applications, including automatic indexing of documents or photographs, the dematerialization of paper-based documents, assistance to visually impaired people, etc..

Prerequisites

Skills in image processing and pattern recognition are necessary for the realization of this internship. The candidate should be familiar with the C + + language. The work to be performed must be integrated into Olena, the free software platform for image processing of the LRDE written in C + +.

Objectives

The context is the analysis of video sequences containing text elements. Once localized, it may be necessary to remove the text following the desired application. The goal of this internship is to remove text in video sequences, especially subtitles. The video images without the text will then be reconstructed thanks to the technique of video inpainting.

Benefit for the candidate
References

1. A. Mosleh, N. Bouguila, and A. Ben Hamza, “Automatic Inpainting Scheme for Video Text Detection and Removal,” IEEE Transactions on Image Processing, vol. 22, no. 11, pp. 4460–4472, 2013.

2. M.Wang,B.Yan,andK.NgiNgan,“Anefficientframeworkforimage/video inpainting,” Signal Processing : Image Communication, vol. 28, no. 7, pp. 753–762, 2013.

3. S. Choi, B. Ham, K. Sohn, “Space-Time Hole Filling With Random Walks in View Extrapolation for 3D Video,” IEEE Transactions on Image Pro- cessing, vol. 22, no. 6, pp. 2429–2441, 2013.

Place LRDE: How to get to us
Compensation

800 € gross/month

Future work opportunities

If you have performed the internship satisfactorily, we would like it to be followed by a PhD thesis.

Contact

jonathan.fabrizio-at-lrde.epita.fr