Temporal Scalability through Adaptive -Band Filter Banks for Robust H.264/MPEG-4 AVC Video Coding
- PDF / 1,129,397 Bytes
- 11 Pages / 600.03 x 792 pts Page_size
- 117 Downloads / 185 Views
Temporal Scalability through Adaptive M-Band Filter Banks for Robust H.264/MPEG-4 AVC Video Coding C. Bergeron,1 C. Lamy-Bergot,1 G. Pau,2 and B. Pesquet-Popescu2 1 EDS/SPM, 2 TSI
THALES Communications, 92704 Colombes Cedex, France Department, Ecole Nationale Sup´erieure des T´el´ecommunications, 75634 Paris Cedex 13, France
Received 15 March 2005; Revised 4 September 2005; Accepted 19 September 2005 This paper presents different structures that use adaptive M-band hierarchical filter banks for temporal scalability. Open-loop and closed-loop configurations are introduced and illustrated using existing video codecs. In particular, it is shown that the H.264/MPEG-4 AVC codec allows us to introduce scalability by frame shuffling operations, thus keeping backward compatibility with the standard. The large set of shuffling patterns introduced here can be exploited to adapt the encoding process to the video content features, as well as to the user equipment and transmission channel characteristics. Furthermore, simulation results show that this scalability is obtained with no degradation in terms of subjective and objective quality in error-free environments, while in error-prone channels the scalable versions provide increased robustness. Copyright © 2006 C. Bergeron et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1.
INTRODUCTION
Modern wireless communication applications relying on the use of video services and video streaming are facing a problem that high-speed wired networks seemed to have overcome: for them, the available bandwidth is still a limiting factor. Moreover, IP wireless networks have to cope with both bit errors and packet losses. This is why a new generation of standards, such as H.264/MPEG-4 AVC finalized in May 2003 [1] jointly by ISO MPEG and ITU-T, and also the new wavelet-based codecs solutions proposed within the scalable video coding (SVC) group, such as [2], take into account the interaction with the network (for the former, through the network abstraction layer concept). Such codecs provide significant compression efficiency improvement when compared to the other existing standards such as MPEG-2 or MPEG-4; and that is why they are so attractive for multimedia applications over wireless communication links. However, H.264/MPEG-4 AVC does not support scalability, which is a very efficient tool to adapt to the bandwidth variations and to the error-prone nature of the wireless channels. Temporal scalability can be achieved using B frames in profiles that support them, which is not the case of H.264/MPEG-4 AVC baseline profile. Solutions are currently being proposed in the literature and within the SVC standardisation group to address this limitation, generally by introducing modifications to the H.264/MPEG-4 AVC syntax to integrate pro-
gressive fine granular scalability coding or subband decompositions [3, 4]. In parallel, solutio
Data Loading...
