Video Block Cross-Component Coding Under Geometric Transformations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in improving coding efficiency and quality, particularly in handling geometric transformations across different color components in video blocks.
Innovation Solution
A method is proposed that involves obtaining information about geometric transformations on video blocks, selecting samples in a first color component to code samples in a second color component, and performing conversions based on these samples to adapt the cross-component coding scheme.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a conventional cross-component coding scheme is used without adapting to geometric transformations, then the coding process is simple, but coding efficiency and quality cannot be improved when geometric transformations are applied
Solution Approach 1:
The patent applies dynamics by making the cross-component coding scheme adaptable to different geometric transformations. The coding parameters and sample selection methods are dynamically adjusted based on the detected transformation type (e.g., rotation, scaling, shearing) rather than using a fixed conventional scheme, thereby improving coding efficiency while managing complexity through conditional adaptation
2Adaptability or versatility
If geometric transformations are applied to video blocks, then video processing flexibility is improved, but coding quality and efficiency deteriorate due to mismatches in cross-component sampling
Solution Approach 1:
The patent applies local quality by selecting and utilizing samples from specific regions and color components based on the geometric transformation characteristics. Different sampling strategies are applied to different color components (e.g., luma vs. chroma) and different spatial regions within the video block, optimizing the coding quality for each local area while maintaining overall transformation flexibility
Data Source
AI summary
Embodiments of the present disclosure provide a solution for video processing. A method for video processing is proposed. The method comprises: obtaining, for a conversion between a current video block of a video and a bitstream of the video, information regarding applying a geometric transformation on the current video block; selecting, based on the information, a set of samples in a first color component of the current video block, the set of samples being used for coding at least one sample in a second color component of the current video block, the second color component being different from the first color component; and performing the conversion based on the set of samples