Spatially Varying Residue Coding for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression standards lack flexibility in residue coding, as they use fixed spatial support transforms, which limits encoding efficiency and visual quality, especially with varying residue characteristics and larger block sizes.
Innovation Solution
Implementing spatially varying residue coding methods that allow multiple transforms within a block and alternate encoding techniques for different parts of the block, along with pre- and post-filtering to improve efficiency and visual quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed transform is used for the entire block, then the encoding process is simple, but the flexibility to encode the residue is significantly reduced
Solution Approach 1:
The block is divided into multiple sub-blocks, and different transforms can be applied to different sub-blocks based on the local residue characteristics. This segmentation allows the encoder to adapt to spatial variations in the residue while maintaining manageable complexity by processing each sub-block independently.
Solution Approach 2:
Different transform types or parameters are applied to different regions (sub-blocks) within the block based on local residue characteristics. This local adaptation improves encoding flexibility and efficiency by matching the transform to the local content properties rather than using a uniform transform for the entire block.
2Productivity
If only one transform is applied to a block, then the encoding is straightforward, but the visual quality and encoding efficiency are limited
Solution Approach 1:
The transform selection becomes dynamic rather than static. Multiple transform options are available, and the encoder can select the most appropriate transform for each sub-block based on local residue characteristics. This dynamic adaptation improves encoding efficiency by optimizing the transform choice for each region.
Solution Approach 2:
Different transform parameters (such as transform type, size, or orientation) are applied to different sub-blocks based on local residue characteristics. This parameter variation allows the encoding process to adapt to local content properties, improving both encoding efficiency and visual quality.
3Adaptability or versatility
If the residue is coded using transform coefficients for the entire block, then all pixels are transformed, but the spatial support of the transforms is fixed reducing flexibility
Solution Approach 1:
The block is segmented into sub-blocks with different spatial supports for the transforms. This allows the encoder to apply transforms with appropriate spatial support to different regions, improving flexibility while maintaining complete residue coding through the combination of all sub-block transformations.
Solution Approach 2:
The spatial support of transforms is varied across different dimensions or regions of the block. By allowing different transform spatial supports in different sub-blocks, the system gains flexibility in adapting to local residue characteristics while ensuring complete coverage through comprehensive sub-block processing.
Data Source
AI summary
Methods and apparatus are provided for spatially varying residue coding. An apparatus includes a video encoder (300) for encoding picture data for at least a block in a picture by selecting a spatially varying encoding method for a residue corresponding to a difference between an original version of the block and at least one reference block. One or more transforms are used to transform a portion of the block and a remaining area of the block is encoded using an alternate encoding method with respect to the one or more transforms or is not encoded.