Back to results
US 9,432,682 B2Granted
System and method for intracoding video data
AssigneeIntel Corporation
InventorGregory J. Conklin
Life of this patent
Estimating the term…
Abstract
A video system for coding a stream of video data that includes a stream of video frames divides each video frame into a matrix of a plurality of subblocks, wherein each subblock includes a plurality of pixels. The video system operates in accordance with nine prediction modes. Each prediction mode determines a prediction mode according to which a present subblock is to be coded. One of the nine prediction modes is selected to encode the present subblock, wherein the selected prediction mode provides for a minimum error value in the present subblock.
Claims24
A method of decoding video data in a computing system, the method comprising:
- decoding a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame, a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, and a value of an upper right-hand corner pixel of a horizontally adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(I+2Q+A+2)/4
- wherein I comprises the value of the upper right-hand corner pixel of the horizontally adjacent subblock, wherein Q comprises the value of the lower right-hand corner pixel of the upper left left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
A non-transitory computer-readable medium including one or more instructions that, when executed by a machine, cause the machine to:
- decode a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame, a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, and a value of an upper right-hand corner pixel of a horizontally adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(I+2Q+A+2)/4
- wherein I comprises the value of the upper right-hand corner pixel of the horizontally adjacent subblock, wherein Q comprises the value of the lower right-hand corner pixel of the upper left left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
An apparatus, comprising:
- circuitry to receive at least a portion of a bitstream of video data;
- and
- circuitry to decode a present subblock of a video frame of the video data according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame, a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, and a value of an upper right-hand corner pixel of a horizontally adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(I+2Q+A+2)/4
- wherein I comprises the value of the upper right-hand corner pixel of the horizontally adjacent subblock, wherein Q comprises the value of the lower right-hand corner pixel of the upper left left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
2 more elements
A non-transitory computer-readable medium including one or more instructions that, when executed by a machine, cause the machine to:
- encode a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame, a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, and a value of an upper right-hand corner pixel of a horizontally adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(I+2Q+A+2)/4
- wherein I comprises the value of the upper right-hand corner pixel of the horizontally adjacent subblock, wherein Q comprises the value of the lower right-hand corner pixel of the upper left left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
An apparatus, comprising:
- circuitry configured to encode a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame, a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, and a value of an upper right-hand corner pixel of a horizontally adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(I+2Q+A+2)/4
- wherein I comprises the value of the upper right-hand corner pixel of the horizontally adjacent subblock, wherein Q comprises the value of the lower right-hand corner pixel of the upper left left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock;
- and
- circuitry configured to generate a bitstream, the bitstream including information indicating the diagonal intra prediction mode.
2 more elements
A method of decoding video data in a computing system, the method comprising:
- decoding a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame and a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, wherein the diagonal intra prediction mode further specifies the wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(Q+A+1)/2
- wherein Q comprises the value of the lower right-hand corner pixel of the upper left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
A non-transitory computer-readable medium including one or more instructions that, when executed by a machine, cause the machine to:
- decode a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame and a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(Q+A+1)/2
- wherein Q comprises the value of the lower right-hand corner pixel of the upper left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
An apparatus, comprising:
- circuitry to receive at least a portion of a bitstream of video data;
- and
- circuitry to decode a present subblock of a video frame of the video data according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame and a value of an upper right-hand corner pixel of a horizontally adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(Q+A+1)/2
- wherein Q comprises the value of the lower right-hand corner pixel of the upper left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
2 more elements
A non-transitory computer-readable medium including one or more instructions that, when executed by a machine, cause the machine to:
- encode a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame and a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(Q+A+1)/2
- wherein Q comprises the value of the lower right-hand corner pixel of the upper left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock.
An apparatus, comprising:
- circuitry configured to encode a present subblock of a video frame according to a diagonal intra prediction mode, wherein the diagonal intra prediction mode specifies a predicted value of an upper left-hand corner pixel of the present subblock based at least in part on a weighted average of a plurality of pixel values, wherein the plurality of pixel values includes a value of a lower right-hand corner pixel of an upper left-hand diagonally adjacent subblock of the video frame and a value of a lower left-hand corner pixel of a vertically adjacent subblock of the video frame, wherein the diagonal intra prediction mode specifies the predicted value P(a) of the upper left-hand corner pixel of the present subblock according to:
- P(a)=(Q+A+1)/2
- wherein Q comprises the value of the lower right-hand corner pixel of the upper left-hand diagonally adjacent subblock, and wherein A comprises the value of the lower left-hand corner pixel of the vertically adjacent subblock;
- and
- circuitry configured to generate a bitstream, the bitstream including information indicating the diagonal intra prediction mode.
2 more elements
Dependencies are inferred from the claim text.