Distributed Video Coding and Processing

A new paradigm in video coding for new applications

Project Description 

Digital video coding is a vital element in many TV and video applications today. A major example is Digital TV, which has also been a driving factor in development of the highly efficient video coders of today, as the MPEG video coding standards. These are characterized by an asymmetry in terms of complexity, typically having one complex encoder and many simpler decoders, which matches broadcast or down-link applications. The current proliferation of video devices and up-link broadband connections has given a radically new picture. In a number of emerging applications e.g. related to wireless communication, the complex encoder is disadvantageous in terms of physical size and power consumption.

The goal of this research is to develop low complexity video encoding, which is still very efficient. These objectives may be facilitated by a paradigm shift in video coding with simple encoding but complex decoding. The new paradigm is referred to as Distributed Video Coding (DVC). It is based on classic results from Information Theory, the Slepian-Wolf and Wyner-Ziv Theorems, regarding independent coding of correlated sources, but retaining performance by joint decoding of the distributed encoded data. Besides providing great benefits in a number of current and emerging applications, this would also enable novel applications of video in communication systems, where a simple encoder and sender communicates with a more powerful (stationary) decoder and receiver or via transcoding at a base station communicates with other simple terminals. An important example is mobile units communicating with base stations and/or servers. Video in sensor networks is another example of an emerging application, which will benefit.

Currently we are working on a drone video system, which can also benefit from the DVC approach. We have started work on multiview DVC and are now also working on combined distributed coding and compressed sensing of multiview images and video.

For more information, you may contact project leader, Prof. Søren Forchhammer.

We have e.g. enhanced the basic TDWZ (Transform Domain Wyner-Ziv) DVC as in Discover to achieve state-of-the-art best in class results by

  • Applying global optical flow in the decoder side motion estimation
  • Iterative generation of side-information and reconstruction
  • Overlapping Block Motion Compensation (OBMC) for generating side information
  • ML-based noise parameter estimation
  • Cross-band noise modeling and refinement

Selected papers:

Luong, Huynh Van; Raket, Lars Lau; Forchhammer, Søren, Re-estimation of Motion and Reconstruction for Distributed Video Coding, IEEE Transactions on Image Processing (DOI: http://dx.doi.org/10.1109/TIP.2014.2320364), vol: 23, issue: 7, pages: 2804-2819, 2014.

Huynh Van Luong, Forchhammer, Søren, Jürgen Slowack, Jan De Cock, Rik Van de Walle, Adaptive Mode Decision with Residual Motion Compensation for Distributed Video Coding, APSIPA Trans. Signal and Inf. Proc, Jan. 2015,

Salmistraro, Matteo; Ascenso, Joao; Brites, Catarina; Forchhammer, Søren, A robust fusion method for multiview distributed video coding, Eurasip Journal on Advances in Signal Processing (DOI: http://dx.doi.org/10.1186/1687-6180-2014-174), vol: 174, 2014.

H. Van Luong, L. L. Raket, X. Huang, and Forchhammer, S., “Side Information and Noise Learning for Distributed Video Coding Using Optical Flow and Clustering,” IEEE Trans. IMAGE Process., vol. 21, no. 12, pp. 4782–4796, Dec. 2012.

Salmistraro, Matteo; Larsen, Knud J.; Forchhammer, Søren, Rate-adaptive BCH codes for distributed source coding, Eurasip Journal on Advances in Signal Processing (DOI: http://dx.doi.org/10.1186/1687-6180-2013-166), vol: 2013, 2013.

Huang, Xin; Forchhammer, Søren, Cross-band noise model refinement for transform domain Wyner–Ziv video coding,Signal Processing: Image Communication (DOI: http://dx.doi.org/10.1016/j.image.2011.06.008), vol: 27, issue: 1, pages: 16-30, 2012


Søren Forchhammer
Professor, Group Leader
DTU Fotonik
+4545 25 36 22