Coding Method Band Offset Sub-Block Division
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional band offset methods in image compression encoding, such as H.265/HEVC, are limited in improving image quality when large CTU block sizes are used, and this limitation is expected to worsen in next-generation standards with even larger CTU sizes.
Innovation Solution
The proposed solution involves dividing the CTU into multiple sub-blocks, determining and calculating band positions for at least two initial sub-blocks, and then applying a band offset to each sub-block using these determined positions, thereby enhancing image quality by adjusting pixel values across regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional band offset methods are used with large CTU block sizes, then device complexity is reduced, but image quality deteriorates
Solution Approach 1:
The CTU block is divided into multiple sub-blocks, and band offset is applied independently to each sub-block. This segmentation allows the band offset process to adapt to local variations within large blocks, improving image quality while maintaining manageable complexity through systematic processing of smaller units.
Solution Approach 2:
Different band offset parameters are applied to different sub-blocks based on their local characteristics. By determining band positions for at least two first sub-blocks and calculating positions for second sub-blocks based on positional relationships, the method adapts the band offset process to local image features, thereby improving overall image quality.
2Productivity
If band offset is applied to the entire CTU block, then processing efficiency is maintained, but precision of pixel value adjustment decreases
Solution Approach 1:
The CTU block is divided into multiple sub-blocks for independent band offset processing. This segmentation enables more precise pixel value adjustment in each sub-block while maintaining overall processing efficiency through systematic handling of divided regions.
Solution Approach 2:
Band offset is applied to all sub-blocks, but with varying degrees of customization - full parameter determination for first sub-blocks and calculated parameters for second sub-blocks. This partial customization approach balances precision requirements with processing efficiency.
3Manufacturing precision
If band positions are determined for all sub-blocks independently, then image quality improves, but calculation complexity increases
Solution Approach 1:
Band positions are determined in advance for at least two first sub-blocks, and then these determined positions are used as reference to calculate band positions for second sub-blocks. This preliminary determination approach reduces calculation complexity by avoiding independent determination for all sub-blocks while maintaining sufficient precision.
Solution Approach 2:
The band positions of first sub-blocks serve as intermediary reference values for calculating the band positions of second sub-blocks. This intermediary approach allows precise band offset application to all sub-blocks while reducing overall calculation complexity through reference-based derivation.
Data Source
AI summary
A coding method by a computer is disclosed. The computer divides a block of a coding unit into a plurality of sub-blocks. The computer determines band positions of at least two first sub-blocks among the plurality of sub-blocks. The computer calculates among the plurality of sub-blocks, band positions of second sub-blocks other than the at least two first sub-blocks based on a positional relationship of the second sub-blocks with respect to the at least two first sub-blocks and the determined band positions for the at least two first sub-blocks in the block of the coding unit. A band offset is conducted for each of the sub-blocks by using the band positions determined or calculated for the at least two first sub-blocks and the second sub-blocks.