4×4 Secondary Transforms for Lower Video Coding Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently processing next-generation video content with high spatial resolution, high frame rate, and high dimensionality, requiring increased memory storage, memory access rates, and processing power, particularly in terms of coding efficiency and complexity.
Innovation Solution
A method is introduced to partition a transform coefficient block into 4×4 blocks and apply individual secondary transforms to each divided block, allowing for sharing of secondary transforms between blocks of varying sizes and shapes, and conditionally applying these transforms based on the number of non-zero coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a transform is applied to high spatial resolution video content, then coding efficiency is improved, but processing complexity increases
Solution Approach 1:
The transform coefficient block is divided into multiple 4×4 sub-blocks, and the secondary transform is applied individually to each sub-block. This segmentation allows the transform to be processed in smaller, more manageable units, improving coding efficiency while controlling processing complexity through localized operations.
Solution Approach 2:
Different secondary transforms can be applied to different 4×4 sub-blocks based on local characteristics such as the number of non-zero coefficients. This local adaptation optimizes coding efficiency for each region while avoiding unnecessary processing in areas where it would not benefit, thus managing overall complexity.
2Manufacturing precision
If secondary transform is applied to entire transform coefficient block, then coding efficiency is improved, but computational complexity increases
Solution Approach 1:
The transform coefficient block is segmented into 4×4 sub-blocks for individual secondary transform processing. This reduces the computational burden on each processing unit while maintaining overall coding efficiency through distributed transformation across multiple smaller blocks.
Solution Approach 2:
The secondary transform is conditionally applied to only those 4×4 sub-blocks that meet specific criteria (e.g., having sufficient non-zero coefficients). This partial application approach maintains coding efficiency for regions where it is beneficial while avoiding unnecessary computational complexity in regions where it would not improve performance.
Data Source
AI summary
Techniques and systems for reconstructing a video signal, which include: obtaining a transform coefficient block by performing an entropy decoding and a dequantization for a current block; deriving a secondary transform corresponding to a specific area in the transform coefficient block, wherein the specific area represents an area including a top-left block of the transform coefficient block; performing an inverse secondary transform for each of subblocks within the specific area using the secondary transform; performing an inverse primary transform for a block which the inverse secondary transform is applied to; and reconstructing the current block using a block which the primary inverse transform is applied to.