A method of padding inactive region for sphere segmented projection of 360 video

摘要

In recent years, 360 videos have been attracting increasing attention as a new type of media that provides immersive experience. In JVET which is exploring next generation of video coding standard with capability beyond HEVC, 360 videos are included in the scope of standard along with SDR and HDR video. In the workflow of 360 video coding of JVET, firstly the 360 videos are projected onto the 2D plane with diverse projection formats, such as equirectangular projection (ERP), cubemap projection (CMP), segmented sphere projection (SSP), rotated sphere projection (RSP), etc. Some projection formats have inactive region in the converted 2D plane such as sphere segmented projection. The inactive region may cause the visual artifact as well as reducing the coding efficiency due to discontinuity at inactive pixel boundary such as pole segments in SSP. This paper proposes a method of padding inactive region to improve coding efficiency and reduce visual artifact. The proposed method of padding inactive region gives -0.2%, -0.2%, - 0.2%, and -0.2% average coding gain of BD-rate in terms of end-to-end S-PSNR-NN, WS-PSNR, the codec level S-PSNR-NN, and WS-PSNR, respectively. In addition, the visual artifacts along the edges between discontinuous faces could be reduced.