Video Coding Transform Coefficient Derivation Using Zero-Out Blocks
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 transmission and storage costs due to increased data volume, and existing compression techniques struggle to efficiently compress and transmit immersive media like virtual reality content.
Innovation Solution
The proposed method enhances image coding efficiency by deriving transform coefficients for a current block based on residual information, using a zero-out block to indicate regions with significant coefficients, and applying multiple transform selection (MTS) to optimize coding. This approach reduces data loss during high-frequency zeroing and improves residual coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution and high-quality images/videos are transmitted, then image quality is improved, but transmission cost and storage cost increase due to increased data volume
Solution Approach 1:
The patent extracts and processes only the significant transform coefficients (non-zero elements) while discarding or approximating the less significant ones through high-frequency zeroing. This extraction approach reduces the effective data volume while preserving the most important image information, directly addressing the contradiction between maintaining image quality and reducing data volume.
Solution Approach 2:
The patent changes the parameter of transform coefficient processing by applying different treatment to different frequency components. Through high-frequency zeroing, low-frequency coefficients are preserved while high-frequency coefficients are set to zero, effectively changing the frequency distribution parameter to reduce data volume while maintaining perceived image quality.
2Productivity
If high-frequency zeroing is applied to reduce data volume, then compression efficiency is improved, but data loss increases
Solution Approach 1:
The patent applies local quality by differentiating the treatment of transform coefficients based on their frequency characteristics. Low-frequency coefficients (which contain most image information) are preserved with high precision, while high-frequency coefficients (which contribute less to perceived quality) are zeroed out. This localized differentiation optimizes the balance between compression efficiency and data loss.
Solution Approach 2:
The patent converts the potentially harmful effect of data loss into a beneficial compression mechanism by strategically accepting loss in high-frequency details while preserving low-frequency information. This approach transforms the trade-off from a problem into a solution, using controlled data loss to achieve significant compression efficiency gains.
3Productivity
If multiple transform selection is used to optimize coding, then residual coding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent applies dynamics by making the transform selection adaptive rather than fixed. The system dynamically chooses the optimal transform type (e.g., DCT, DST, ADCT) based on the specific characteristics of each block, allowing the coding system to adjust its behavior according to content requirements. This dynamic adaptation improves residual coding efficiency while managing complexity through intelligent selection rather than implementing all transforms simultaneously.
Data Source
AI summary
A video decoding method according to the present document comprises a step of deriving transform coefficients for a current block on the basis of residual information, wherein the step of deriving the transform coefficients comprises a step of deriving a zero-out block indicating a region in which effective transform coefficients may exist in the current block, wherein the zero-out block is derived on the basis of flag information indicating whether multiple transform selection (MTS), in which a plurality of transform kernels are used, can be applied to the current block.