Hybrid interframe coding of video signals with backward-acting motion detection

Motion compensation can be performed by either "forward-acting" or "backward-acting" motion detection. Forward algorithms require the transmission of motion vectors, whereas backward algorithms estimate the motion from two previously reconstructed frames at both the transmitter and receiver ends. For this reason, the backward systems are more attractive to low-rate videoconferencing applications. However, backward motion estimation may suffer some degradation due to varying speed of moving objects. In this paper we investigate the tradeoffs between the two schemes of motion compensation.