LM Mode Chroma Prediction Simplification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression techniques, particularly in the Linear Model (LM) mode of High-Efficiency Video Coding (HEVC), face complexity issues due to large bit-width multipliers, which hinder coding efficiency and performance.
Innovation Solution
The introduction of new parameters such as avgY′, avgC, RErrY′, and RErrC, which reduce the bit widths of multipliers and simplify calculations, allowing for improved intra prediction in the LM mode by using rounding or flooring functions to compute averages and differences of neighboring samples.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard LM mode calculation is used, then prediction accuracy is maintained, but computational complexity increases due to large bit-width multipliers
Solution Approach 1:
The patent segments the complex LM mode calculation into distinct functional modules: a linear model prediction unit that computes preliminary predictions, a gradient compensation unit that calculates corrections, and a combination unit that merges results. This modular segmentation reduces computational complexity by allowing each unit to operate with simplified calculations while maintaining overall prediction accuracy through the coordinated function of all units.
Solution Approach 2:
The patent changes the parameters used in calculation by introducing gradient-based compensation terms that modify the standard LM mode formula. Instead of using large bit-width multipliers directly, the patent transforms the calculation into gradient computations with smaller multipliers, thereby reducing device complexity while preserving prediction precision through the mathematical equivalence of the transformed parameters.
2Productivity
If compression ratio is increased to reduce data quantity, then transmission efficiency improves, but implementation complexity increases
Solution Approach 1:
The patent replaces complex mechanical computation systems with optimized arithmetic operations. By substituting the standard multi-step LM mode calculation with a streamlined gradient-based approach using smaller multipliers, the patent reduces implementation complexity while maintaining the high compression ratio needed for efficient transmission.
Solution Approach 2:
The patent transforms the compression algorithm parameters by using gradient computations instead of direct large-multiplier operations. This parameter change enables the system to achieve high compression ratios with reduced computational overhead, thereby improving transmission efficiency without proportionally increasing implementation complexity.
3Productivity
If encoding speed is increased to reduce processing time, then productivity improves, but coding performance deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-computing gradient values and storing them in lookup tables or caching mechanisms. This allows the encoding process to quickly retrieve and apply gradient compensations without performing complex real-time calculations, thereby increasing encoding speed while maintaining coding performance through the use of pre-prepared accurate gradient data.
Solution Approach 2:
The patent replaces time-consuming mechanical computation with faster arithmetic operations and memory access. By substituting complex multi-step calculations with streamlined gradient-based computations and lookup table retrievals, the patent achieves higher encoding speeds without sacrificing coding performance, as the simplified operations are mathematically equivalent to the original complex calculations.
Data Source
AI summary
A video codec comprising a processor computing an average of a plurality of reconstructed chroma samples located in neighboring blocks of a chroma block and adjacent to the chroma block, computing an intermediate variable by applying an integer function on the average of the reconstructed chroma samples, and generating a predicted chroma sample for the chroma block based on the intermediate variable.