Fast CU size decision algorithm using machine learning for HEVC intra coding

作者：

Highlights：

•

• The proposed algorithm is to reduce encoding time of the HEVC.

• Using adaptive online learning phase, the thresholds are updated automatically.

• The CU size is early determined by the FLDA and k-NN.

• The proposed algorithm successfully reduce the encoding time of the HEVC.

摘要

•The proposed algorithm is to reduce encoding time of the HEVC.•Using adaptive online learning phase, the thresholds are updated automatically.•The CU size is early determined by the FLDA and k-NN.•The proposed algorithm successfully reduce the encoding time of the HEVC.

论文关键词：HEVC,Fast coding unit size decision,Fisher’s linear discriminant analysis, k-nearest neighbors classifier

论文评审过程：Received 12 June 2017, Revised 11 December 2017, Accepted 11 December 2017, Available online 16 December 2017, Version of Record 21 December 2017.

论文官网地址：https://doi.org/10.1016/j.image.2017.12.005