Complexity modeling for coarse grain scalable (CGS) video decoding

This paper proposes a hybrid model to predict CGS-SVC decoding complexity. We take advantage of both the statistic characteristic of complexity features and linear relationship between quality layers to model the complexity. Experimental results show that the proposed method provides a good prediction accuracy for all quality layer. The whole average prediction error of test sequences is 1.51% approximately. Furthermore, the target platform can decode the suitable quality layer by our layer decision mechanism and an accurate prediction result.