Method and system for coding and decoding information associated with video compression

摘要

A method and system for coding and decoding information associated with video impression is described. The video sequence is processed in a plurality of frames. Each frame of the plurality of frames is processed in a plurality of macroblocks. A prediction of an original video signal, which is a part of a macroblock, in a current frame, is constructed from the video sequence. A residual signal is formed by subtracting the prediction of the original video signal from the original video signal in the current frame. A transform to the residual signal is applied. A plurality of transform coefficients is quantized. A symbol of at least one syntax element that defines a characteristic of the residual signal is identified. Symbols of the at least one syntax element of a same category are coded together.

主权项

1. A method of video compression, the method comprising:
encoding macroblocks of a sequence of video frames; generating coded block patterns for the macroblocks, wherein the coded block patterns include bits that indicate whether different video blocks of the macroblocks include non-zero transform coefficients; and run-length coding the coded block patterns for the macroblocks together such that one or more runs defined in run-length coding span symbols of a plurality of different coded block patterns associated with a plurality of the macroblocks, and wherein the coded block patterns comprise luma coded block patterns, and wherein run-length coding the coded block patterns comprises run-length coding the luma coded block patterns together such that one or more first runs span symbols of a plurality of different luma coded block patterns associated with the plurality of the macroblocks.