Hardware-based parallel context-based estimation for video encoding
AssigneeNVIDIA Corporation
InventorsYonghai WU, Jianjun CHEN, Jinchang XU, Yang ZHANG, Yongmao TANG, Wei FENG
Life of this patent
Estimating the term…
Loading patent…
AssigneeNVIDIA Corporation
InventorsYonghai WU, Jianjun CHEN, Jinchang XU, Yang ZHANG, Yongmao TANG, Wei FENG
Estimating the term…
Various embodiments include techniques for encoding a media frame. A video encoder encodes media frames in a video stream by dividing the media frames into blocks of pixels and encoding numerical coefficients associated with the blocks to generate a bit stream. As part of the encoding process, the video encoder estimates the bit rate needed to encode a set of blocks of the media frame. The video encoder generates more accurate bit rates for encoding the blocks by basing the bit rate estimation on various context-based parameters of the coefficients rather than just on the coefficients themselves. This technique for context-based bit rate estimation can improve the visual quality of the encoded video stream. Further, the video encoder encodes blocks using various simplifying assumptions to eliminate interblock dependencies without unduly reducing visual quality. As a result, each block can be encoded independently and in parallel with encoding other blocks.
A computer-implemented method for encoding a first block of a current media frame, the method comprising:
A system comprising:
Dependencies are inferred from the claim text.