Video compression method for man-machine mixed vision

• Main Authors: YI XIAOKAI, WANG HANLI
• Format: Patent
• Language: Chinese; English
• Published: 09.12.2022
• Subjects: ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; PICTORIAL COMMUNICATION, e.g. TELEVISION

The invention relates to a video compression method for man-machine mixed vision, which comprises the following steps that: 1) a video compression network acquires a video sequence to be compressed, encodes the video sequence through an encoder, and compresses the video sequence into an encoding code stream; 2) the coding code stream enters a downstream task processing network, if the coding code stream faces human vision requirements, the step 3) is executed, and if the coding code stream is used for machine vision analysis, the step 4) is executed; (3) a decoder in the human vision branch module receives the encoding code stream and decodes the encoding code stream into a video, and the video is output after being enhanced; and 4) a decoder in the machine vision branch module receives the encoding code stream, decodes the encoding code stream to generate a video, analyzes the video according to the vision task, and outputs an analysis result. Compared with the prior art, the method can meet the requirements