Decoding methods, video decoders, and computer-readable storage media

Abstract

The invention relates to a method for decoding video, a video decoder, and a computer-based storage medium. The computer-based device performs a method for decoding video data by obtaining a video bit stream consisting of data combined with encoded images, each image comprising rows of coding units (CTU), each CTU comprising one or more coding units (CU). The data buffer stores multiple motion vector predictions based on history used to encode the rows of CTUs, and the decoding process resets the buffer before decoding the first Cl ĩ of the current row of CTUs. For the current current data set of the CTUJs, a list of motion vector candidates is constructed by exploiting the spatial and temporal correlation of the motion vectors of neighboring encoding units, as well as motion vector predictors based on history in the buffer. Finally, a motion vector predictor is selected from the list of motion vector candidates to decode the current data set, and the buffer is updated based on the selected motion vector predictor.