抄録
It is really important to decide proper prediction mode in H.264 encoder, since there much more modes than conventional method such as MPEG-2. H.264 encoder like JSVM uses squared error as the criterion of distortion for mode decision. However, squared error dose not always correspond to the distortion from the viewpoint of subjective quality. In this paper, we investigate the mode decision method based on human visual system in order to improve H.264 encoder.
