An efficient content adaptive transform for video coding

In this paper, we propose a content adaptive transform (CAT) method for video coding. Firstly, CAT obtains the transform kernels in real-time instead of predefined fixed ones, and therefore it is content based transform. Secondly, the transform kernels is obtained by singular value decomposition (SVD) from a typical decoded block without transmitting them to the decoder. Thirdly, the optimality property of CAT for input signal is derived. Finally, a spiral scanning pattern is proposed to rearrange the diagonal distributed transform coefficients. Extensive experimental results show that, compared with the H.264/AVC High Profile, the proposed CAT scheme provides average PSNR gain 0.55 dB for intra frame coding on a large number of multiple-resolutions sequences.