Rate - Distortion - Efficient Video Compression Using a 3 - D

In this paper we combine model-based video synthesis with block-based motion-compensated prediction. Two frames are utilized for prediction where one frame is the previously decoded one and the other frame is provided by a model-based coder. The approach is integrated into an H.263-based video codec. Rate-distortion optimization is employed for the coding control. Hence, the coding ee-ciency does not decrease below H.263 even if the model-based coder cannot describe the current scene. On the other hand, if the objects in the scene correspond to the model-based coder, signiicant gains in coding eeciency can be obtained compared to TMN-10, the test model of the H.263 standard. This is veriied by experiments with natural head-and-shoulder sequences. Bit-rate savings of about 35 % are achieved at equal average PSNR. When encoding at equal bit-rate, signiicant improvements in terms of subjective quality are visible.