Hybrid Transforms in Video Coding Resolving Size Mismatch
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding techniques face inefficiencies in reducing data size, particularly when transform block sizes differ from prediction block sizes, leading to suboptimal coding efficiency.
Innovation Solution
Implementing hybrid transforms with different transform modes for vertical and horizontal pixels within transform subblocks, allowing for adaptive encoding that matches the proximity of each subblock to prediction edges, thereby optimizing data compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single transform mode is used for the entire residual block, then the encoding process is simple, but coding efficiency is suboptimal when transform block size differs from prediction block size
Solution Approach 1:
The residual block is divided into multiple transform subblocks, each of which can be transformed using different transform modes. This segmentation allows the encoder to adapt to local characteristics within the residual block, improving coding efficiency when transform block size differs from prediction block size while maintaining manageable encoding complexity through systematic processing of subblocks.
Solution Approach 2:
Different transform modes (vertical, horizontal, or diagonal) are applied to different transform subblocks based on their local characteristics and proximity to prediction edges. This local adaptation optimizes the transformation for each region's specific residual patterns, thereby improving overall coding efficiency without requiring complex global optimization.
2Productivity
If transform block size is made different from prediction block size to adapt to residual patterns, then coding efficiency improves, but the encoding complexity increases
Solution Approach 1:
By segmenting the residual block into transform subblocks, the system can apply different transform modes to different regions, effectively adapting to varying residual patterns without requiring a complete redesign of the encoding process for the entire block. This reduces the complexity increment compared to a fully adaptive approach.
Solution Approach 2:
The patent applies transform modes selectively to transform subblocks rather than requiring full adaptation across the entire residual block. This partial action approach achieves sufficient adaptation to residual patterns while avoiding the excessive complexity that would result from completely adaptive transformation throughout the entire block.
Data Source
AI summary
A block of video data can be encoded using intra prediction followed by transforming the generated residual block where the transform size is different from the prediction size. A plurality of transform modes for the residual block is provided whereby the transform subblocks of the residual block are transformed using horizontal and vertical one-dimensional transform types. The transform types may be selected such that their base function corresponds to a pattern in the data of the generated residual block resulting from the intra prediction mode. As a result, the position of each block relative to the peripheral pixels used to generate a prediction block for the block may be used to select the transform types.