Video Decoding Sub-Block Transform Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The sub-block transform (SBT) technology in AVS3 can result in excessively narrow sub-blocks, degrading hardware performance and affecting coding and decoding efficiency.
Innovation Solution
By determining a partitioning mode for coding blocks based on their size and an allowable maximum sub-block size ratio, the proposed solution avoids the creation of excessively narrow sub-blocks during SBT, ensuring hardware performance and improving coding and decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If sub-block transform (SBT) technology is applied to partition coding blocks, then coding precision and detail representation are improved, but hardware performance degrades due to excessively narrow sub-blocks
Solution Approach 1:
The patent introduces a minimum sub-block size parameter (minSubBlockWidth and minSubBlockHeight) to constrain the partitioning process. This parameter change ensures that sub-blocks maintain a minimum dimensional threshold, preventing excessively narrow sub-blocks that would degrade hardware performance while still allowing sufficient partitioning for coding precision.
Solution Approach 2:
The patent implements dynamic partitioning control where the available partitioning modes are determined based on the coding block size and the minimum sub-block size constraint. The system adaptively selects from multiple partitioning modes (vertical half, vertical quad, horizontal half, horizontal quad) while ensuring resulting sub-blocks meet the minimum size requirement, balancing coding precision with hardware performance.
2Productivity
If coding blocks are partitioned into narrower sub-blocks to improve coding efficiency, then coding and decoding speed increase, but hardware performance deteriorates
Solution Approach 1:
By introducing minimum sub-block size parameters (minSubBlockWidth and minSubBlockHeight), the patent establishes lower bounds on sub-block dimensions. This parameter constraint allows the system to partition blocks sufficiently to maintain coding efficiency while preventing the creation of excessively narrow sub-blocks that would harm hardware performance.
Solution Approach 2:
The patent segments the coding block partitioning process into multiple stages with controlled segmentation ratios. Instead of allowing unlimited fine-grained partitioning, the system divides blocks into manageable segments (half or quad partitions) that respect the minimum size constraint, achieving a balance between coding efficiency and hardware performance.
3Manufacturing precision
If maximum sub-block size ratio is increased to allow finer partitioning, then coding precision improves, but the number of excessively narrow sub-blocks increases
Solution Approach 1:
The patent introduces minimum sub-block size parameters (minSubBlockWidth and minSubBlockHeight) that act as lower bounds on sub-block dimensions. These parameter changes prevent the generation of excessively narrow sub-blocks even when the maximum sub-block size ratio allows for fine partitioning, thereby eliminating the harmful effect while preserving coding precision.
Solution Approach 2:
The patent converts the potential harm of excessive partitioning (creating too many narrow sub-blocks) into a benefit by using the minimum size constraint to guide the partitioning process. The constraint ensures that only beneficial partitions producing adequately sized sub-blocks are executed, transforming the partitioning control mechanism into a protective filter against harmful fine-grained division.
Data Source
AI summary
A video encoding method includes: acquiring a coding block of a video image frame; determining a partitioning mode available to the coding block according to a size of the coding block and an allowable maximum sub-block size ratio in response to a determination that sub-block transform (SBT) is to be performed on the coding block; determining, based on the partitioning mode available to the coding block, a target partitioning mode used during the SBT performed on the coding block; and partitioning the coding block based on the target partitioning mode, and decoding a plurality of sub-blocks obtained through the partitioning.