An Adaptive Nested Scalar Quantization Scheme for Distributed Video Coding

The quantizer plays an important role not only in traditional video compression standards but also in the Wyner-Ziv coding, which can be thought to consist of a quantizer followed by a Slepian-Wolf encoder. In this context, this paper proposes a frame-level adaptive nested scalar quantization scheme for distributed video coding (DVC). At the encoder the absolute frame difference between current Wyner-Ziv frame and previous key frame is utilized to denote the statistical dependency, and then the adaptive quantizer stpesize is decided according to threshold and sent to decoder. The Wyner-Ziv frame is decoded and reconstructed based on the received parameter and the side information. It is shown that at a possible cost of higher error rate of channel code, the proposed adaptive quantization scheme is a potential way to lower the quantization error and on the whole improve the rate distortion performance, especially for video with low motion.