Spatially Varying Transforms for Residual Block Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently compressing video data to support high-definition streaming without prolonged waiting times, particularly in adapting transforms based on the spatial distribution of residual blocks.
Innovation Solution
Implementing a spatial varying transform (SVT) that positions transform blocks at various candidate locations relative to residual blocks, using different transforms based on these positions, including SVT vertical (SVT-V) and SVT horizontal (SVT-H) types, and employing inverse Discrete Sine Transform (DST) or Discrete Cosine Transform (DCT) based on block boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a fixed transform is applied to all residual blocks, then the processing is simple and fast, but the coding efficiency is suboptimal for different spatial distributions of residual blocks
Solution Approach 1:
The patent applies dynamics by making the transform type adaptive rather than fixed. The encoder dynamically selects between different transform types (e.g., DST, DCT) based on the spatial distribution characteristics of each residual block, allowing the system to optimize coding efficiency for different content patterns while maintaining a manageable complexity through predefined transform options
Solution Approach 2:
The patent changes the transform parameter (transform type) based on residual block characteristics. By analyzing the spatial distribution of non-zero coefficients in residual blocks and adjusting the transform type accordingly, the system optimizes compression efficiency without requiring complex custom transforms for each scenario
2Productivity
If multiple transform types are applied based on residual block position, then coding efficiency improves, but the complexity of determining and applying the correct transform increases
Solution Approach 1:
The patent segments the residual block processing into different regions or patterns based on spatial distribution characteristics. By dividing the analysis into distinct categories (e.g., based on position of non-zero coefficients), the system can apply appropriate transforms to each segment, improving overall efficiency while managing complexity through systematic classification
Solution Approach 2:
The patent performs preliminary analysis of residual block characteristics before transform application. By pre-determining the spatial distribution pattern and selecting the optimal transform type in advance, the system avoids complex real-time decisions during encoding, thereby improving compression efficiency while controlling processing complexity
3Adaptability or versatility
If transform blocks are positioned at various candidate locations, then adaptability to different residual patterns improves, but the complexity of positioning and selecting transforms increases
Solution Approach 1:
The patent applies local quality by allowing different transform types to be applied to different spatial regions or patterns within the residual block. Each region is analyzed for its specific characteristics, and the transform is customized accordingly, enabling high adaptability to various residual patterns while managing complexity through localized processing
Data Source
AI summary
A mechanism for position dependent spatial varying transform (SVT) for video coding. A prediction block and a corresponding transformed residual block are received at a decoder. A type of spatial varying transform (SVT) employed to generate the transformed residual block is determined. A position of the SVT relative to the transformed residual block is also determined. An inverse of the SVT is applied to the transformed residual block to reconstruct a reconstructed residual block. The reconstructed residual block is then combined with the prediction block to reconstruct an image block.