Method for 360 video processing based on multiple viewpoints and apparatus therefor

作者: Lee Sooyeon , Oh Sejin

DOI:

关键词: ViewpointsMetadataVideo processingDecoding methodsArtificial intelligenceRendering (computer graphics)Multiple viewpointData processingComputer visionViewportComputer science

摘要: A method for 360 video data processing based on multiple viewpoints performed by a receiving apparatus according to the present invention comprises steps of: acquiring, from received signals, image information pictures of viewpoints-based videos and metadata videos, wherein includes viewpoint information; decoding an first viewport basis rendering into 3D space; second deriving, in case which hotspot is selected, connected through space.