Allocation adaptative des bits et techniques de sélection de résolution pour l'estimation du mouvement dans un codeur vidéo à ondelettes

Wavelet transform is an important tool for image and video coding applications. Several motion estimation techniques have been proposed in the wavelet domain. Coarse-to-fine motion estimation techniques generally have a lower complexity, but they may lead to inaccurate motion estimation if the coarse motion estimation is erroneous. On the other hand, pixel-domain motion estimation techniques provide a superior estimation, but at a higher complexity. In this paper, an efficient video coder in the wavelet domain is proposed. First, an adjustable resolution selection for motion estimation is proposed where a low-pass subband at an appropriate resolution is employed for coarse motion estimation. The motion vectors corresponding to subbands of other resolutions are then predicted from these coarse motion vectors. When motion vectors are predicted from a lower- to a higher-resolution subband, the predicted motion vectors are further refined using a small search window. Second, an adaptive bit allocation technique is proposed where the bits are allocated optimally between the motion vectors and the displaced frame difference. This allocation is performed by minimizing a cost function based on the Lagrange multiplier method. Simulation results show that the proposed video coder provides a good coding performance compared to other multiresolution techniques proposed in recent literature.