Self-learning residual model for fast intra CU size decision in 3D-HEVC

作者：

Highlights：

• Residual signal is exploited as the feature of each CU, which is strongly correlated with the optimal CU partition.

• A self-learning residual model is established by iteratively learning from the features of the previously encoded coding tree unit (CTU).

• A binary classifier is developed with the self-learning residual model for early CU size decision in 3D-HEVC.

摘要

•Residual signal is exploited as the feature of each CU, which is strongly correlated with the optimal CU partition.•A self-learning residual model is established by iteratively learning from the features of the previously encoded coding tree unit (CTU).•A binary classifier is developed with the self-learning residual model for early CU size decision in 3D-HEVC.

论文关键词：3D high efficiency video coding,Self-learning,Fast intra coding,CU size decision,Residual signal

论文评审过程：Received 3 April 2019, Revised 2 August 2019, Accepted 3 October 2019, Available online 8 October 2019, Version of Record 9 October 2019.

论文官网地址：https://doi.org/10.1016/j.image.2019.115660