Fast ISP Mode Decision for the Versatile Video Coding Intra Prediction Using Machine Learning

The Versatile Video Coding (VVC) standard achieves high compression rates by introducing new encoding tools, such as the Intra Subpartition Prediction (ISP). However, the ISP increases the computational effort necessary to perform the mode decision of the intra prediction step. This paper proposes a fast intra-mode decision solution for the ISP using machine learning. A Decision Tree is employed to predict the most promising ISP modes to be optimal to avoid the costly RDO test of ISP modes that are less likely to be chosen. By reducing the number of modes fully evaluated by the RDO process, the proposed solution achieves an average time-saving of 3.15% with only 0.11% of coding efficiency loss when tested for the common test conditions of VVC. Unlike the related works, our solution avoids the time overhead of calculating image features by adopting features from the encoding process. Compared with related works, our solution presents competitive time-saving and coding efficiency results.