Back to results
US 12,707,078 B1Granted
Video frame affine motion compensation
AssigneeAmazon Technologies, Inc.
InventorsChristopher Andrew Segall, Kiran Mukesh Misra, Byeongdoo Choi
Life of this patent
Estimating the term…
Abstract
Frames may be encoded at different resolutions. A scaling factor may be used to scale video frames used for translational or affine motion compensation. Various improvements to motion compensation processes for different resolution video frames are described.
Claims20
A method, comprising:
- obtaining video content and two or more control point motion vectors between a block of samples in a prediction frame of the video content and a previous frame of the video content that has been encoded and decoded;
- determining a first ratio of the previous frame to the prediction frame in a first dimension and second ratio of the previous frame to the predicted frame in a second dimension, wherein either the first ratio, the second ratio, or both are greater than 1;
- identifying a block of samples in the previous frame using a location of the block of samples in the prediction frame, the control point motion vectors, the ratio in the first dimension, and the ratio in the second dimension;
- and
- coding the block of samples in the prediction frame based on the block of samples in the previous frame using an affine motion compensation process, wherein non-adjacent samples of the block of samples in the previous frame are used to determine motion compensated pixel values for the block of samples in the prediction frame, wherein the non-adjacent samples are selected based on the first ratio and the second ratio, wherein the affine motion compensation process comprises generating an intermediate block based on the second block, the first ratio, and the second ratio, wherein the intermediate block is not included in a coded bitstream.
2 more elements
A method, comprising:
- obtaining video content;
- and
- coding a first block of a first frame of the video content based on a second block of a second frame of the video content using an affine motion compensation process, wherein a ratio of the resolution of the second frame to the resolution of the first frame is greater than 1, wherein the affine motion compensation process comprises generating an intermediate block based on the second block and the ratio, wherein the intermediate block is not included in a coded bitstream.
A system, comprising one or more processors and one or more memories, the one or more memories storing computer-executable instructions that, when executed by the one or more processors, causes the one or more processors to perform operations including:
- obtaining a bitstream representing encoded video content;
- and
- decoding a first block of a first frame of the video content based on a second block of a second frame of the video content using an affine motion compensation process, wherein a ratio of the resolution of the second frame to the resolution of the first frame is greater than 1, wherein the affine motion compensation process comprises generating an intermediate block based on the second block and the ratio, wherein the intermediate block is not included in a coded bitstream.
Dependencies are inferred from the claim text.