< Back to previous page

Publication

Side-information dependent correlation channel estimation in hash-based distributed video coding

Journal Contribution - Journal Article

In the context of low-cost video encoding, distributed video coding (DVC) has recently emerged as a potential candidate for uplink-oriented applications. This paper builds on a correlation channel modeling concept which expresses the correlation noise as being statistically dependent on the side-information. Compared to classical side-information independent noise model-ing adopted in current DVC solutions, it is theoretically proven that side-information dependent modeling improves the Wyner-Ziv coding performance. Anchored in this finding, this paper proposes a novel algorithm for online estimation of the side-information dependent correlation channel parameters based on already decoded information. The proposed algorithm enables bit-plane-by-bit-plane successive refinement of the channel esti-mation leading to progressively improved accuracy. Additionally, the proposed algorithm is included in a novel DVC architecture which employs a competitive hash-based motion estimation tech-nique to generate high quality side-information at the decoder. Experimental results corroborate our theoretical gains and vali-date the accuracy of the channel estimation algorithm. The per-formance assessment of the proposed architecture shows remark-able and consistent coding gains over a germane group of state-of-the-art distributed and standard video codecs, even under strenuous conditions, that is, large groups of pictures (GOP) and highly irregular motion content.
Journal:  IEEE Trans Image Process
ISSN: 1057-7149
Issue: April 2012
Volume: 21
Pages: 1934-1949
Publication year:2012
Keywords:distributed video coding, hash information, overlapped block motion estimation, correlation channel, online successively refined channel estimation, video coding
  • ORCID: /0000-0001-7290-0428/work/84065550
  • ORCID: /0000-0001-9300-5860/work/71094993
  • ORCID: /0000-0003-0908-1655/work/69212838
  • WoS Id: 000302181800041
  • Scopus Id: 84859025948