Distributed Video Coding: Identifying Promising Application Scenarios

Distributed Source Coding (DSC) is a new coding paradigm based on two major Information Theory results: the Slepian-Wolf [1] and Wyner-Ziv Theorems [2,3]. DSC theory relies on the coding of two or more dependent random sequences in an independent way, i.e. associating an independent encoder to each sequence. A single decoder is used to perform joint decoding of all encoded sequences, exploiting the statistical dependencies between them. Based on the DSC independent encoding-joint decoding configuration, a new video coding paradigm, called Distributed Video Coding (DVC) has been defined. Although the theoretical foundations of distributed source coding have been established in the 1970s, the design of practical video coding schemes based on DSC has been proposed only in recent years [4-13]. A major reason behind these developments is related to the evolution of channel coding, notably the emergence of turbo and LDPC (low-density parity-check) codes, which provide ways to build the channel codes necessary for DVC. The major objective of this MPEG contribution is to report on the study made by the European project DISCOVER about the application scenarios for which the DVC paradigm may bring major benefits and identify which are these benefits. Note it is not the purpose of this contribution to claim that DVC is the right way to go for any application scenario. Considering the far from mature stage of DVC research, it is too early for final conclusions and claims. The purpose is rather to identify the most promising applications, helping the researchers to focus their work on the most adequate application spots, in order conclusions on the value of DVC for these applications may be taken as soon as possible. The benefits discussed along this document are valid under the assumption that the major objectives of the DISCOVER project (e.g. flexible allocation of complexity, low encoding complexity, increased error robustness) may be reached, at least within an acceptable degree. Although the literature generally refers that DVC is useful for low complexity and low-power consumption encoders, no detailed application analysis is available on these advantages. It is also believed by the DISCOVER consortium that low complexity is not the single potential DVC advantage, and may not even be the most promising one. This investigation is precisely one of the major research targets of this project. To achieve the objectives stated above, this contribution identifies and studies which requirements and functionalities are relevant for each application scenario, for example, coding efficiency, error resilience, and encoder-decoder complexity trade-off. This contribution also clusters the application scenarios according to various relevant characteristics, e.g. single/multiple cameras, encoder/decoder complexity, delay constraints, availability of return channel. Finally, a list with the application scenarios for which DVC looks to be more promising will be drawn, following a proposed methodology. [1] J. Slepian and J. Wolf, “Noiseless Coding of Correlated Information Sources”, IEEE Trans. on Information Theory, vol. 19, nº 4, pp. 471 - 480, July 1973; [2] A. Wyner, “Recent Results in the Shannon Theory”, IEEE Trans. on Information Theory, vol. 20, nº 1, pp. 2 - 10, January 1974; [3] A. Wyner and J. Ziv, “The Rate-Distortion Function for Source Coding with Side Information at the Decoder”, IEEE Trans. on Information Theory, vol. 22, nº 1, pp. 1 - 10,January 1976; [4] R. Puri and K. Ramchandran, “PRISM: A New Robust Video Coding Architecture Based on Distributed Compression Principles”, 40th Allerton Conference on Communication, Control and Computing, Allerton, USA, October 2002; [5] A. Majumdar and K. Ramchandran, “PRISM: an Error-Resilient Video Coding Paradigm for Wireless Networks”, First International Conference on Broadband Networks, San Jose, CA, USA October 2004, 2004 Page(s):478 – 485; [6] A. Majumdar, R. Puri, P. Ishwar and K. Ramchandran, “Complexity/Performance Tradeoffs for Robust Distributed Video Coding”, IEEE Int. Conference on Image Processing, Genova, Italy, September 2005; [7] A. Aaron, R. Zhang and B. Girod, “Wyner-Ziv Coding of Motion Video”, Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, USA, November 2002; [8] A. Aaron, S. Rane, E. Setton and B. Girod, “Transform-Domain Wyner-Ziv Codec for Video”, Visual Communications and Image Processing Conference, San Jose, CA, USA, January 2004; [9] B. Girod, A. Aaron, S. Rane and D. Rebollo Monedero, “Distributed Video Coding”, Proceedings of the IEEE, vol. 93, nº 1, pp. 71 - 83, January 2005; [10] A. Aaron, S. Rane and B. Girod, “Wyner-Ziv Video Coding with Hash-Based Motion Compensation at the Receiver”, IEEE Int. Conference on Image Processing, Singapore, October 2004; [11] Abhik Majumdar, Jiajun Wang, and Kannan Ramchandran, “Drift Reduction in Predictive Video Transmission using a Distributed Source Coded Side-Channel”, ACM Multimedia, October 2004; [12] S. Rane, A. Aaron and B. Girod, “Systematic Lossy Forward Error Protection for Error-Resilient Digital Video Broadcasting - a Wyner-Ziv Coding Approach”, IEEE International Conference on Image Processing, Singapore, October 2004; [13] A. Sehgal and N. Ahuja, “Robust Predictive Coding and the Wyner-Ziv Problem”, Data Compression Conference, Snowbird, Utah, USA, October 2003.