MSc Semester Project
Title
Error Resilience for Scalable Distributed Video Coding
Candidate
Megha Agarwal
Supervisor
Prof. Touradj Ebrahimi
Assistant: Mourad Ouaret
Expected
2008, summer semester
Place
EPFL
Description
Distributed source coding (DSC) is a new coding paradigm based on two Information Theory theorems: Slepian-Wolf and Wyner-Ziv. Theoretically, it states that the optimal rate achieved when performing joint encoding and decoding of two or more correlated sources can be reached by doing separate encoding and joint decoding. In other words, a single decoder is used to perform joint decoding exploiting the statistical dependencies. Based on DSC, a new video coding paradigm, referred to as distributed video coding (DVC) is defined.
In this project, we propose to implement scalable video coding schemes based on DVC. Then, these schemes need to be evaluated in a realistic scenario.
The student will have to perform the following tasks:
- Study DVC and state of the art of scalable DVC.
- Several scalable schemes for video based on DVC were introduced in [1]. The student needs to deeply study these schemes and integrate them into the DVC software.
- Study the influence of the base layer bit allocation on the performance of the scalable schemes.
- Study how the conventional and WZ streams should be packetized and what is the appropriate packet size to use.
[1] M. Ouaret, F. Dufaux, and T. Ebrahimi. Codec-Independent Scalable Distributed Video Coding. In 2007 IEEE International Conference on Image Processing (ICIP2007), 2007.