Dual-layer approach for improved mesh-based motion compensation
Andy C. Yu, H. Park, G.R. Martin · IET Image Processing · 2007
A dual-layer algorithm for mesh-based motion compensation is described. The technique utilises meshes at different scales and aims to improve the rate-distortion performance at fixed bit rates. The proposed algorithm identifies motion-active regions in the picture without the need for time-consuming evaluation. By enhancing the mesh structure in the motion-active regions, a significant improvement in motion estimation is obtained. Furthermore, a scheme to construct two different significance maps, at both encoder and decoder, provides a reduction in transmission overhead. A block-wise motion information coding approach is proposed, permitting a group of static grid points to be represented by a one-bit codeword. Simulations are reported for a wide range of test sequences possessing different degrees of motion activity. The results demonstrate that the proposed algorithm achieves a better PSNR-rate performance compared with conventional mesh-based motion compensation methods.