Directional Non-Separable Transform for Inter-Coded Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques lack non-separable transforms for inter-coded blocks, limiting compression efficiency and computational complexity reduction.
Innovation Solution
Implement non-separable transforms, such as low frequency non-separable transforms (LFNST), for inter-coded blocks, which are direction-dependent and applied based on block characteristics like gradients and geometric partitioning modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If non-separable transforms are applied to inter-coded blocks, then compression efficiency is improved, but device complexity increases
Solution Approach 1:
The non-separable transform is segmented into two independent one-dimensional transforms: a first transform applied in a first direction and a second transform applied in a second direction. This segmentation allows the complex two-dimensional non-separable transform to be decomposed into simpler, more manageable operations that can be implemented efficiently while maintaining the compression benefits.
Solution Approach 2:
The patent applies transforms in multiple directions (first direction and second direction) rather than only in traditional separable horizontal and vertical directions. This dimensional approach enables the transform to capture directional characteristics of video data more effectively, improving compression efficiency by better aligning with the actual content structure.
2Manufacturing precision
If direction-dependent non-separable transforms are applied, then video coding performance is improved, but computational complexity increases
Solution Approach 1:
The transform kernel is made adaptive and direction-dependent, allowing it to dynamically adjust based on the characteristics of the video block. The system determines appropriate transform directions and applies corresponding kernels that match the content, optimizing coding performance for each specific case rather than using a fixed transform for all blocks.
Solution Approach 2:
The patent changes the parameters of the transform operation by selecting different transform directions and corresponding kernels based on block characteristics. This parameter adaptation allows the system to optimize the transform for each specific video block, improving coding performance while managing computational complexity through selective application.
Data Source
AI summary
Systems and techniques are described for processing video data. For example, an apparatus can obtain a block of video data encoded using inter-prediction. In some examples, the apparatus can determine a direction associated with the block of video data, for instance based on pixels of the block of video data, based on pixels of at least one neighboring block of the block of video data, or based on information associated with a geometric partitioning mode (GPM) associated with the block of video data. The apparatus can apply a non-separable transform to the block of video data, for instance to decode or encode the video data. In some examples, the apparatus can apply the non-separable transform to the block according to the direction associated with the block of video data.