Learning-based CU partition prediction for fast panoramic video intra coding

Wenjun Zheng, Chao Yang, Ping An, Xinpeng Huang, Liquan Shen

Abstract: Highlights•We propose a learning-based fast intra-partition algorithm for 360°video coding.•We propose to utilize four SVM-based classifiers to accelerate CU partition in VVC.•We assign different QP values to different regions in encoding 360°video frames.•Experiments and analyses are conducted to show the excellence of our method.