Towards A Multi-Terminal Video Compression Algorithm By Integrating Distributed Source Coding With Geometrical Constraints

In this paper, we present a framework for multiterminal video compression (MTVC) that exploits the geometric constraints between cameras with overlapping fields of view, and then uses distributed source coding on corresponding points in two or more views. There are two main parts of the paper a Distributed Motion Estimation (DME) algorithm and a Distributed Source Coding (DSC) algorithm. The novelty of the approach lies in the integration of these two parts. The overall algorithm is developed such that future improvements in either of the two parts can be integrated into it. We start by describing the algorithm for obtaining corresponding points followed by the distributed coding strategy. Our experiments provide a benchmark on the upper bound of the gain obtained through distributed coding of a pair of video sequences, as opposed to coding each of them separately. We also show results on a reallife face video sequence obtained by a pair of overlapping cameras. Through this experimental analysis, we show that at low bit-rates MTVC can achieve very significant savings.