A Novel Fast Multiple Reference Frames Selection Method for H.264/AVC

Multiple reference frames motion estimation is one of the new features introduced in H.264/AVC. Although this new feature improves video coding performance, it is extremely computational intensive. Full search scheme is adopted in reference software JM, and the increased computation load is in proportion to the number of searched reference frames. In this paper, we propose a novel fast multiple reference frames selection method. It can decrease the number of reference frames for motion estimation and reduce the complexity of coding adaptively according to the features of video sequences. The experiment results show that our algorithm achieves 42% coding time saving on average with negligible quality loss.