Suboptimal prediction for Motion Compensated Temporal Filtering in Scalable Video Coding

This paper presents the way of ´suboptimal´ prediction for motion compensated temporal filtering (MCTF) in scalable video coding (SVC). MCTF is one of the major techniques for temporal scaling in SVC that consists of prediction and update steps. Adaptive prediction approaches have been studied to derive high quality images in lower layer, however it requires large computational cost. We develop a novel approach of determining prediction filter coefficients in suboptimal sense that achieves high image quality at low computational cost. The proposed method is evaluated through computer simulation in comparison with some conventional techniques.