Image Coding Using LFNST and Intra DC Mode Update
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and high-quality images/videos, such as 4K and 8K ultra high definition, leads to higher bit rates, resulting in increased transmission and storage costs. Additionally, the need for efficient compression techniques is exacerbated by the rise of immersive media like virtual and augmented reality.
Innovation Solution
The proposed method involves an image coding technique that uses a transform-based approach to enhance coding efficiency. Specifically, it updates the intra prediction mode of chroma blocks to an intra DC mode based on the intra prediction mode of the corresponding luma block, determines an LFNST set comprising LFNST matrices, and performs an LFNST on the chroma block using the derived LFNST matrix.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution and high-quality images/videos (4K, 8K UHD) are transmitted or stored using conventional methods, then image quality is improved, but transmission cost and storage cost increase
Solution Approach 1:
The patent applies transform techniques (including LFNST - Low Frequency Non-Separable Transform) to change the representation parameters of image data from spatial domain to frequency domain, enabling more efficient compression. By transforming residual signals and applying optimized transform matrices, the patent achieves better energy compaction and reduced bitrate while maintaining high image quality
Solution Approach 2:
The patent replaces conventional transform mechanisms with optimized transform techniques including LFNST and RST (Reduced Secondary Transform). These substituted transform methods provide improved compression efficiency by better capturing the correlation structure in transform coefficients, thereby reducing the bitrate required for high-quality image transmission and storage
2Loss of energy
If transform techniques are applied to compress image data, then transmission and storage costs are reduced, but coding complexity increases
Solution Approach 1:
The patent segments the transform process into distinct stages: primary transform, secondary transform (including LFNST and RST), and quantization. By dividing the complex transform operation into manageable segments with specific functions, the patent reduces overall coding complexity while maintaining compression efficiency
Solution Approach 2:
The patent extracts and applies optimized transform techniques specifically to residual signals and transform coefficients. By taking out only the necessary transform operations (LFNST/RST) and applying them selectively based on block characteristics, the patent reduces unnecessary computational complexity while achieving effective compression
3Productivity
If LFNST is applied to chroma blocks, then compression efficiency is improved, but determination of LFNST set increases processing complexity
Solution Approach 1:
The patent performs preliminary determination of the LFNST set based on the intra prediction mode of the corresponding luma block before applying LFNST to chroma blocks. By pre-determining which LFNST matrices to use based on already-available luma prediction mode information, the patent avoids additional complex decision-making processes and reduces processing complexity
Solution Approach 2:
The patent uses the intra prediction mode information from the luma block (which is already processed) to automatically determine the LFNST set for chroma blocks. This self-service approach leverages existing data without requiring separate complex analysis, thereby improving compression efficiency while minimizing additional processing complexity
Data Source
AI summary
An image decoding method according to the present document comprises the steps of: on the basis of an intra prediction mode of a chroma block being a cross-component linear model (CCLM) mode and an intra prediction mode of a luma block corresponding to the chroma block being a palette mode, updating the intra prediction mode of the chroma block to an intra DC mode; on the basis of the updated intra prediction mode, determining an LFNST set including LFNST matrices; and performing LFNST on the chroma block on the basis of an LFNST matrix derived from the LFNST set, wherein the intra DC mode is an intra prediction mode corresponding to a specific location within the luma block.