Fast adaptive loop filter algorithm for high efficiency video coding

Adaptive loop filter (ALF) is one of the new tools adopted in the next generation video coding standard HEVC. ALF improves both BD Rate and subjective performance by using Wiener filtering. However, the computational complexity of ALF is quite high. ALF accounts for 14% of the total encoder runtime. In this paper, two fast ALF algorithms, LCU based edge sub-sample scheme and Patch based edge sub-sample scheme, are proposed. The implementation result shows that only 0.04% and 0.16% BD Rate drop with 40% and 71% runtime is saved for LCU based edge sub-sample scheme and Patch based edge sub-sample scheme respectively.