Low complexity multiple description coding method for networked video

Abstract

Robust transmission of compressed digital video over the best-effort internet and wireless networks is a proven challenge due to packet loss, and multiple description (MD) coding is an effective way to cope with the problem. In this paper, we propose a new MD video coding method. It starts from GOB alternation to create two descriptions for each frame encoded. By carrying motion information in both the descriptions, the quality of single-channel reconstruction is greatly improved. A process we call it low-quality macroblock update helps to further improve the quality of single-channel reconstruction and reduce the difference between single-channel and two-channel reconstruction, this in turn reduces the mismatch between the MD video encoder and decoder. The proposed method has very good performance both in terms of redundancy rate distortion and in the packet loss scenario, with relatively lower complexity. It is standard compliant to the extent that the output bit streams (descriptions) can be decoded by a standard video decoder. A remarkable feature of the method is that it can be applied to nearly all block-based video codec.