Optimized State-Distortion Extraction for Scalable Extension of H.264/AVC

A GOP-based state extraction algorithm for scalable video coding (SVC) is proposed to improve the video streaming transmission efficiency and reconstructed quality. According to the scalable features and internal relationship of SVC enhancement layers, we put forward a novel state model, based on which the SVC bit streams extraction algorithm employing the rate-distortion optimization method is presented to obtain the optimal enhancement layers. By using the state model, the rate-distortion feature of different SVC enhancement layers is analyzed conveniently, and the enhancement layers owning minimum distortion is selected efficiently, which ensures the transmission priority for the enhancement layer owning the best rate-distortion performance with certain bandwidth limitations. The proposed algorithm takes fully consideration of the rate-distortion performance of various enhancement layers in different states, and improves the reconstructed quality of video sequence. The simulation results show the proposed method achieves better performance.