Intersection motion vector prediction in video encoding

Abstract

Example computer-readable methods, devices, and media are described. An example device includes one or more processors configured to determine, for each of a plurality of subblocks of a current block of a current video data image, a list of corresponding intersection motion vectors. Each list of corresponding intersection motion vectors includes at least one intersection motion vector. The at least one intersection motion vector includes a source motion vector associated with a first reference image and a target motion vector associated with a second reference image. A source block is located in the first reference image and has a predictor that is a subset of a target block located in the second reference image.One or more processors are configured to determine an intersection move vector for the current block based on the corresponding intersection move vector lists and decode the current block based on the intersection move vector.