The JCT-VC standardized Screen Content Coding (SCC) extension in the HEVC HMRExt + SCM reference codec offers an impressive coding efficiency performancewhen compared with HM RExt alone; however, it is not significantly perceptuallyoptimized. For instance, it does not include advanced HVS-based perceptualcoding methods, such as JND-based spatiotemporal masking schemes. In thispaper, we propose a novel JND-based perceptual video coding technique for HMRExt + SCM. The proposed method is designed to further improve the compressionperformance of HM RExt + SCM when applied to YCbCr 4:4:4 SC video data. In theproposed technique, luminance masking and chrominance masking are exploited toperceptually adjust the Quantization Step Size (QStep) at the Coding Block (CB)level. Compared with HM RExt 16.10 + SCM 8.0, the proposed method considerablyreduces bitrates (Kbps), with a maximum reduction of 48.3%. In addition tothis, the subjective evaluations reveal that SC-PAQ achieves visually losslesscoding at very low bitrates.
展开▼
