Drifting Reduction in Progressive Fine Granularity Scalable Video Coding

This paper investigates the drifting problem in Progressive Fine Granularity Scalable (PFGS) video coding, which arises from the fact that high quality references are used to predict the enhancement layer. Due to the unique property of the PFGS scheme, this paper first analyzes the occurrence and propagation of the drifting errors in details. As a result, an iterative model is established to simulate the drifting errors at the encoder. With the proposed drifting model, a decisionmaking mechanism can help the PFGS scheme to choose the coding mode of each macroblock in the enhancement layer. The experimental results show that the PFGS scheme incorporating the proposed technique can effectively reduce the drifting errors at low bit rates. Meanwhile, the PFGS scheme also improves the coding efficiency up to 1.8dB compared with the FGS scheme at moderate or high bit rates.