Chroma Bi-Prediction Weighting From Reconstructed Luma Blocks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression techniques, such as VVC, face challenges in efficiently setting weights for chroma components during bi-prediction, which affects coding efficiency and video quality.

Innovation Solution

A method for inter-predicting chroma components by calculating weights at the decoder side using previously reconstructed luma prediction blocks or template matching, applied to chroma bi-prediction blocks to generate a final prediction block.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If weights are assigned at CU level for chroma components during bi-prediction, then coding efficiency is improved, but the complexity of weight setting and optimization increases

Engineering Contradiction:
Improvecoding efficiencyVSAvoidweight setting complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent uses luma prediction blocks as an intermediary to derive chroma weights. Instead of directly optimizing chroma weights which is complex, the method leverages the already-computed luma prediction blocks and their weights as a mediator to indirectly determine optimal chroma weights, simplifying the overall weight setting process while maintaining coding efficiency

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the parameter approach by deriving chroma weights from luma domain parameters. The chroma weights are not independently optimized but are derived by transforming luma prediction block characteristics into chroma weight values, reducing the search space and computational complexity for weight optimization

Inventive Principle:
Principle #35Parameter changes

2Reliability

If bi-prediction is performed with CU-level weights for chroma components, then video quality is enhanced, but processing time and computational resources increase

Engineering Contradiction:
Improvevideo qualityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by pre-computing luma prediction blocks and their associated weights during the encoding process. These pre-computed luma weights are then directly utilized for chroma weight derivation, eliminating the need for separate chroma weight computation and reducing overall processing time while maintaining high video quality

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent merges the weight derivation processes for luma and chroma components. Instead of independently computing weights for both components, the method combines the weight computation into a single process where luma weights are derived and then directly applied to chroma prediction, reducing redundant computations and processing time

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS20250301121A1Method for inter-prediction of chroma component using bi-prediction
Publication Date: 2025.09.25 HYUNDAI MOTOR CO LTD
  • US20250301121A1 patent drawing
  • US20250301121A1 patent drawing
  • US20250301121A1 patent drawing

AI summary

A method is disclosed for inter-predicting chroma components by using bi-prediction. A video decoding device generates, for the current chroma block, a chroma bi-prediction block. The video decoding device derives weights of the current chroma block by using a first luma prediction block, a second luma prediction block, and a final luma reconstruction block of a luma block corresponding to the current chroma block. The video decoding device generates a final chroma prediction block of the current chroma block by applying the weights to the chroma bi-prediction blocks.