Google patents video coding trick to save hardware power
This filing describes a method for compressing video by predicting color pixels based on brightness pixels. It limits the number of calculations needed for this prediction based on the size of the video block, which helps hardware process the data more efficiently.
In the filing’s own words
Tap-constrained convolutional cross-component model (CCCM) prediction enables hardware coder implementations of CCCM prediction by limiting the number of taps used to predict chroma samples while maintaining accuracy in the prediction. During encoding, a current luma sample of a block is identified. A number of taps to use for predicting a chroma sample associated with the current luma sample is determined based on a size of the block and/or whether the block is downs ampled. The chroma sample is predicted using a prediction model limited to the number of taps and then encoded to an encoded bitstream. During decoding, a current luma sample of a block and a number of taps for predicting a chroma sample associated with the current luma sample are decoded from an encoded bitstream. The chroma sample is predicted using a prediction model limited to the number of taps and then output within an output video stream.
What exactly they claim (claim 1)
1. A method for tap-constrained convolutional cross-component model prediction, the method comprising: identifying a current luma sample of a block to encode; determining a number of taps to use for predicting a chroma sample associated with the current luma sample, wherein the number of taps is based on one or more of a size of the block or whether the block is downsampled, and wherein taps of the number of taps are convolutional cross-component model prediction filter coefficients; predicting the chroma sample using a prediction model limited to the number of taps; and encoding the current luma sample and the number of taps to an encoded bitstream.
Facts
- Company
- Google (GOOGLE LLC)
- Official title
- TAP-CONSTRAINED CONVOLUTIONAL CROSS-COMPONENT MODEL PREDICTION
- Publication number
- US20260261666A1
- Application number
- 19642127
- Filed
- Apr 8, 2026
- Published
- Sep 3, 2026
- Inventors
- 6
- Claims
- 20
Classified under
H04N 19/117 Electricity & electronicsH04N 19/132 Electricity & electronicsH04N 19/176 Electricity & electronicsH04N 19/186 Electricity & electronicsH04N 19/80 Electricity & electronicsMore from Google in AI & software
- Google patents video search that answers questions using AI descriptions Sep 15, 2026
- Google patents photo tool that removes people and their gear Sep 15, 2026
- Google patents AI that writes live meeting notes and summaries Sep 15, 2026
- Google patents AI tool that turns messy emergency notes into clear messages Sep 3, 2026
- Google patents AI that clicks through apps to finish tasks for you Sep 24, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.