Fast CU Partition and Intra-Mode Decision for 360° Videos Using VVC

360-degree videos (abbreviated as 360 videos) have gained widespread popularity due to their immersive experience. The massive amount of raw data resulting from ultra-high resolution 360 videos makes the coding process extremely complex and seriously limits their widespread applications. In this paper, we propose a new fast Coding Unit (CU) partition and Intra Mode (IM) decision algorithm for 360 videos using Versatile Video Coding (VVC). The major novelty is that we establish databases and develop models based on stretching and distribution for 360 videos using VVC. Specifically, first, we establish stretching-based CU partition and distribution-based IM decision databases for 360 videos. Second, we propose a new stretching-based multi-scale convolution kernel network structure and a composite loss function, which incorporates stretching factors and rate-distortion (RD) costs as well as unbalance issue. We then develop a multi-threshold selection scheme to select candidate Split Modes (SMs). Finally, we design a distribution-based IM prediction network and its corresponding loss function. Experimental results demonstrate that the proposed algorithm improves coding speed by 73.42%, with a marginal Bjøntegaard Delta Bit Rate (BDBR) increase of only 2.34%, thereby outperforming current state-of-the-art methods.