Modeling Correlation Noise Statistics at Decoder for Pixel

Distributed video coding (DVC) is a new video coding paradigm based on two key Information Theory results: the Slepian-Wolf and Wyner-Ziv theorems. Recently, promising results were shown in Wyner-Ziv (WZ) video coding, a particular case of DVC. In the literature, many practical WZ coding approaches model the correlation noise between the original frame and the so-called side information by a given distribution whose relevant parameters are estimated offline, at the encoder. This paper proposes an algorithm to estimate, at the decoder, and at the frame level, the correlation or error distribution between the original and the side information frames, in a way which is as efficient as the estimation made at the encoder based on the original information. This approach relieves the encoder from the task to perform this estimation based on the original information, which is rather important since DVC solutions are typically adopted under low encoder complexity constraints.