Optimal bit allocation in scalable multiple description video coding for packet loss resilience

Abstract

Scalable multiple description video coders (SMDC) allow for better adaptation of video transmission to varying network conditions with compression efficiency comparable to that of non-scalable MDC coders. Recently, we proposed a new SMDC video coding framework, which offers flexibility in the number of descriptions, redundancy level in each description, and bitrate of each description by post encoding processing only [2]. In this work, we propose a low complexity algorithm to find the best low and high rates for this new framework to strike the optimal trade off between compression efficiency and error resilience at a given packet loss rate. We provide experimental results, by means of network simulations, to compare our method with SMDC using a fixed number of descriptions and fixed redundancy level.