Transform Kernel Selection Strategy for the H.264

In this paper, we make use of a multiple-kernel scheme as a guide to exploit the transform kernel selection strategy of the hybrid video coding. We have found that the kernel selection tendency lies in a feature which can be extracted from a pair of kernels. We propose using the IK(1,2,1) kernel for Iand P-Frames and the IK(5,7,3) kernel for B-Frames for the H.264 Video Coding Standard. This gives a good improvement in terms of the PSNR and bitrate compared to those using a single kernel in the H.264 (JM 12.2) and the multiple-kernel schemes available in the literature. We also generalize a feature extracted from a pair of kernels to form a feature extracted from a group of many kernels. This collection of corresponding operation points forms the Macroblock-Level Kernel Galaxy (MLKG). Regarding to the robustness and ease of visualizing the performance, we also propose using the MLKG as a guide to design a single-kernel transform coding process for possible future video coding standards.