Block-based picture fusion for context segmentation and processing

摘要

The encoder receives the video frame, partitions the video frame into blocks, determines a first region within the video frame that includes a first group of a first subset of blocks, and measures a first average of information in the first region. circuitry configured to determine a value and encode the video frame, wherein encoding includes controlling the quantization parameter based on a first average measured value of information in the first region. Related devices, systems, techniques and articles are also described.