A New Intra Coding Method for Spatial Scalability in Scalable Video Coding

In scalable video coding, spatial scalability is a layered video representation with several layers corresponding to different spatial resolutions for the same video sequence. Therefore, spatial correlation between successive two layers exists naturally, and should be further exploited. In this paper, a new scalable intra coding method for spatial scalability is proposed. It consists of two spatial layers: one base layer and one enhancement layer. The proposed intra coding method is fully compatible with H.264/AVC standard in base layer, while in enhancement layer the upsampled reconstructed base layer blocks are used for intra prediction under condition of that the adjacent prediction pixels are not available. The effectiveness of the new technique is demonstrated by experimental results. With almost no changes on both computation complexity and encoded bit-rate, the average increase of Y-PSNR is about 0.13 dB for sequence Football and 0.11 dB for Foreman