Spatial-Temporal Correlation Based Fast Intra Mode Decision for HEVC

Zeqi Feng,Pengyu Liu,Kebin Jia,Kun Duan
2018-01-01
Abstract:In high efficiency video coding (HEVC), up to 35 intra prediction modes are adopted to improve encoding performance. However, large numbers of modes lead to increased coding complexity inevitably, which hinders the promotion of HEVC in real-time environment. To reduce the encoding complexity, a spatial-temporal correlation based fast intra mode decision algorithm is proposed in this paper. Herein, the rough mode decision (RMD) candidate list and the rate distortion optimization (RDO) candidate list are narrowed respectively by strong spatial correlation and temporal correlation between prediction units (PUs). For RMD, the proposed algorithm early determines the impossible range of the PU optimal mode in line with spatial correlation, which would narrow RMD candidate list and speed up RMD process. For RDO, the proposed algorithm tries to combine temporal correlation with intra coding to simplify RDO candidate list and save RDO time consuming. Experimental results demonstrate that the proposed algorithm yields average 28.64% encoding time reduction with 1.19% BDBR gain or 0.06dB BDPSNR loss compared with HM16.9. The proposed algorithm drives HEVC coding efficiency under real-time environments.
What problem does this paper attempt to address?