Uniform Bit-Width Multipliers for LM Mode Chroma Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression algorithms, particularly the Linear Model (LM) mode, face high computational complexity due to the use of multipliers with varying bit widths, which complicates the intra prediction process in video coding.
Innovation Solution
The introduction of a video codec system that computes predicted chroma samples using multipliers with bit widths no greater than the input bit depth of the chroma block, employing flooring functions of averages of neighboring samples to simplify calculations and reduce bit width variability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing algorithms use multipliers with varying bit widths for chroma prediction, then prediction accuracy is maintained, but computational complexity increases
Solution Approach 1:
The patent changes the parameter of multiplier bit width from variable to uniform (capped at input bit depth). By constraining all multipliers to have bit width no greater than the input bit depth of the chroma block, the patent simplifies the computational structure while maintaining prediction accuracy through the reformulated linear model equations.
2Measurement precision
If complex computations are used in LM mode, then prediction quality is improved, but processing time increases
Solution Approach 1:
The patent modifies the computational parameters by limiting multiplier bit widths to the input bit depth, which reduces the number of computational operations required. This parameter change enables faster processing while preserving prediction quality through the mathematically equivalent reformulated equations.
3Device complexity
If uniform bit width multipliers are used, then implementation complexity is reduced, but prediction accuracy may deteriorate
Solution Approach 1:
The patent introduces a parameter transformation that reformulates the linear model equations to work with uniform bit width multipliers. By expressing the prediction in terms of averaged neighboring samples with controlled bit widths, the patent achieves both simplified implementation and maintained accuracy.
Solution Approach 2:
The patent creates a simplified copy of the prediction mechanism that uses uniform bit width multipliers instead of variable bit width multipliers. This copied version maintains functional equivalence while reducing implementation complexity through standardized arithmetic operations.
Data Source
AI summary
A video codec able to generate a predicted chroma sample for a chroma block based on reconstructed luma samples and reconstructed chroma samples. The reconstructed luma samples are located in neighboring blocks of a corresponding reconstructed luma block and adjacent to the corresponding reconstructed luma block. The reconstructed chroma samples are located in neighboring blocks of the chroma block and adjacent to the chroma block. The predicted chroma sample is generated by computing multipliers. Each of the multipliers has a bit width no greater than an input bit depth of the chroma block. The predicted chroma sample is generated without the need for any multiplier having a bit width greater than the input bit depth of the chroma block.