Intensity estimation/compensation for interlaced forward-predicted fields
AssigneeMicrosoft Corporation
InventorsThomas W. Holcomb, Sridhar Srinivasan, Shankar Regunathan
Life of this patent
Estimating the term…
Abstract
Techniques and tools for intensity compensation for interlaced forward-predicted fields are described. For example, a video decoder receives and decodes a variable length code that indicates which of two reference fields for an interlaced forward-predicted field use intensity compensation (e.g., both, only the first, or only the second). The decoder performs intensity compensation on each of the two reference fields that uses intensity compensation. A video encoder performs corresponding intensity estimation/compensation and signaling.
Claims16
A computer-implemented method for transforming encoded video information using a video decoder, the method comprising:
- with the video decoder:
- receiving encoded video information in a bitstream;
- decoding, from the encoded video information in the bitstream, a syntax element indicating an entire interlaced forward-predicted field has two reference fields rather than a single reference field;
- decoding, from the encoded video information in the bitstream, a variable length code that represents information indicating which of the two reference fields for the interlaced forward-predicted field use intensity compensation, wherein the information indicates whether both, only the first, or only the second of the two reference fields use intensity compensation, and wherein the variable length code is signaled at field level for the interlaced forward-predicted field;
- performing intensity compensation on at least part of each of the two reference fields that uses intensity compensation, wherein the performing intensity compensation comprises adjusting plural pixel values in each of the two reference fields that uses intensity compensation based at least in part on intensity compensation parameters;
- and
- performing motion compensation for one or more parts of the interlaced forward-predicted field relative to the two reference fields after the adjustment for the intensity compensation, wherein both, only the first, or only the second of the two reference fields used in the motion compensation has pixel values adjusted for intensity compensation.
A video decoder comprising:
- a processor;
- memory;
- and
- storage storing computer-executable instructions implementing:
- means for decoding a syntax element indicating an entire interlaced forward-predicted field has two reference fields rather than a single reference field;
- means for decoding a variable length code to determine which of the two reference fields use intensity compensation for the interlaced forward-predicted field, wherein the variable length code indicates whether both, only the first, or only the second of the two reference fields use intensity compensation, and wherein the variable length code is signaled at field level for the interlaced forward-predicted field;
- means for performing intensity compensation on at least part of either one or both of the two reference fields that use intensity compensation, wherein the performing intensity compensation comprises adjusting plural pixel values in each of the two reference fields that use intensity compensation based at least in part on intensity compensation parameters;
- and
- means for performing motion compensation for one or more parts of the interlaced forward-predicted field relative to the two reference fields after the adjustment for the intensity compensation, wherein both, only the first, or only the second of the two reference fields used in the motion compensation has pixel values adjusted for intensity compensation.
A computer-implemented method for transforming video information using a video encoder, the method comprising:
- with the video encoder:
- determining that an interlaced forward-predicted field has two reference fields rather than a single reference field;
- signaling, in an output bitstream, a one-bit code indicating that the entire interlaced forward-predicted field has two reference fields;
- determining whether to perform intensity compensation for each of the two reference fields of the interlaced forward-predicted field;
- performing intensity compensation on at least part of each of the two reference fields that are determined to use intensity compensation, wherein the performing intensity compensation comprises adjusting plural pixel values in each of the two reference fields that are determined to use intensity compensation based at least in part on intensity compensation parameters;
- performing motion compensation for one or more parts of the interlaced forward-predicted field relative to the two reference fields after the adjustment for the intensity compensation, wherein both, only the first, or only the second of the two reference fields used in the motion compensation has pixel values adjusted for intensity compensation;
- coding a variable length code that represents information indicating which of the two reference fields for the interlaced forward-predicted field use intensity compensation, wherein the information indicates whether both, only the first, or only the second of the two reference fields use intensity compensation;
- and
- signaling the variable length code in the output bitstream, wherein the variable length code is signaled at field level for the interlaced forward-predicted field.
Dependencies are inferred from the claim text.