Method and system for generating a transform size syntax element for video decoding
View Patent ↗In a video processing system, a method and system for generating a transform size syntax element for video decoding are provided. For high profile mode video decoding operations, the transform sizes may be selected based on the prediction macroblock type and the contents of the macroblock. A set of rules may be utilized to select from a 4.×.4 or an 8.×.8 transform size during the encoding operation. Dynamic selection of transform size may be performed on intra-predicted macroblocks, inter-predicted macroblocks, and/or direct mode inter-predicted macroblocks. The encoding operation may generate a transform size syntax element to indicate the transform size that may be used in reconstructing the encoded macroblock. The transform size syntax element may be transmitted to a decoder as part of the encoded video information bit stream.
1. A method for video signal processing, the method comprising:
receiving, from a video information stream, transform coefficients representing a coded macroblock in an image;
determining a macroblock type of the coded macroblock;
receiving a transform size syntax element from the video information stream, wherein the transform size syntax element indicates a transform size to be used for inverse transforming the transform coefficients in the image;
selecting an inverse transform size based on the received transform size syntax element and a selection rule corresponding to the determined macroblock type, wherein for at least one inter-coded macroblock type the selected inverse transform size does not correspond to a transform size indicated by the received transform size syntax element;
inverse transforming the transform coefficients to determine inverse transformed coefficients based on the selected inverse transform size; and
generating a reconstructed output based on the inverse transformed coefficients.
2. The method of claim 1 , wherein receiving the transform coefficients from the video information stream further comprises:
determining that the transform coefficients represent an inter-coded region of the image.
3. The method of claim 1 ,
wherein the transform size syntax element indicates that the transform size is 4×4, and
wherein selecting the inverse transform size further comprises selecting the 4×4 inverse transform size as the inverse transform size based on the transform size syntax element.
4. The method of claim 3 ,
wherein inverse transforming the transform coefficients further comprises inverse transforming the transform coefficients using a 4×4 spatial prediction and a 4×4 inverse transform.
5. The method of claim 1 ,
wherein the transform size syntax element indicates that the transform size is 8×8, and
wherein selecting the inverse transform size comprises selecting the 8×8 inverse transform size as the inverse transform size based on the transform size syntax element.
6. The method of claim 5 ,
wherein inverse transforming the transform coefficients further comprises inverse transforming the transform coefficients using a 8×8 spatial prediction and a 8×8 inverse transform.
7. The method of claim 1 ,
wherein the transform size syntax element indicates a first transform size;
wherein the determined macroblock type is the at least one inter-coded macroblock type; and
wherein selecting the inverse transform size comprises selecting a second inverse transform size different than the first transform size indicated by the transform size syntax element.