Coefficient Classification Using Texture Information for Wyner-Ziv Video Coding

"What part of the video signal should be coded using what method" is a challenge for Wyner-Ziv video coding (WZVC). The challenge results from that in WZVC the temporal prediction is performed at the decoder, so the encoder hardly determines what part of the signal should be coded using what method. In this paper a novel coefficients classification strategy utilizing both texture information and the difference between the current and previous frame is proposed to determine "what part" should be encoded with priority. Simulations results confirm that the proposed strategy is effective for improving rate distortion of WZVC.