Back to results
US 11,240,492 B2Granted
Neural network based residual coding and prediction for predictive coding
AssigneeApple Inc.
InventorsJiefu Zhai, Xingyu Zhang, Xiaosong Zhou, Jun Xin, Hsi-Jung Wu, Yeping Su
Life of this patent
Estimating the term…
Abstract
Systems and methods disclosed for video compression, utilizing neural networks for predictive video coding. Processes employed combine multiple banks of neural networks with codec system components to carry out the coding and decoding of video data.
Claims16
A method for coding a video stream, comprising:
- for a pixel block of an input frame to be coded, generating a pixel block prediction based on input data derived from reference data of previously-coded data of the video stream, wherein generating the prediction uses a neural-network-based prediction that includes transforming the reference data from a pixel domain to a transform domain;
- generating a residual block representing a difference between the pixel block and the pixel block prediction;
- coding the residual block and
- packing the coded residual block and associated coding parameters in a coded video stream.
1 more element
A method for decoding a coded video stream, comprising:
- for a coded residual block to be decoded, extracting the coded residual block and associated coding parameters from the coded video stream;
- decoding the coded residual block, based on the coding parameters, generating a decoded residual block;
- generating a pixel block prediction based on input data derived from reference data of previously decoded data of the coded video stream, wherein generating the pixel block prediction uses a neural-network-based prediction that includes transforming the reference data from a pixel domain to a transform domain;
- and generating a reconstructed pixel block, the reconstructed pixel block is a sum of the decoded residual block and the pixel block prediction.
1 more element
A computer system, comprising:
- at least one processor;
- at least one memory comprising instructions configured to be executed by the at least one processor to perform a method comprising:
- for a coded residual block to be decoded, extracting the coded residual block and associated coding parameters from the coded video stream wherein the neural-network-based prediction includes transforming the reference data from a pixel domain to a transform domain;
- decoding the coded residual block, based on the coding parameters, generating a decoded residual block;
- generating a pixel block prediction based on input data derived from reference data of previously decoded data of the coded video stream;
- and generating a reconstructed pixel block, the reconstructed pixel block is a sum of the decoded residual block and the pixel block prediction.
3 more elements
A method for coding a video stream, comprising:
- for a pixel block of an input frame to be coded, generating a pixel block prediction, using a neural-network-based prediction, based on input data derived from reference data of previously-coded data of the video stream wherein the neural-network-based prediction includes transforming the reference data from a pixel domain to a transform domain;
- generating a residual block representing a difference between the pixel block and the pixel block prediction;
- coding the residual block and
- packing the coded residual block and associated coding parameters in a coded video stream.
1 more element
Dependencies are inferred from the claim text.