An adaptive framework for H.264 optimized encoding
Rafael Galvão de Oliveira, Maria Trocan, Béatrice Pesquet‐Popescu, Nicolas Tizon · 2011
Lossy compression using subband decomposition and quantization has been proven its efficiency, being part of both image and video coding standards. In this paper, we propose an adaptive encoding framework for H.264/AVC, in which the inverse transformation matrix is optimized as function of both content and quantization noise. The inverse transform matrix is adapted in an iterative manner for each frame, and an efficient coding scheme of the resulting synthesis transform is provided. The proposed transform optimization method shows promising results compared to H.264/AVC.