Adaptive Largest Coding Unit Partitioning for Intra Video Slices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as HEVC, face limitations in efficiently modulating the largest coding unit (LCU) size within a picture, particularly for regions using intra prediction, leading to suboptimal compression efficiency and increased slice header overhead.
Innovation Solution
The proposed method determines the largest coding unit size for a picture and applies partitioning based on whether the LCU is an intra type, allowing for adaptive LCU sizes and shapes within a picture, enabling more efficient coding by containing intra regions in intra slices and inferring QT and BT split flags for intra slices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a fixed largest coding unit size is used for the entire picture, then the coding structure is simple, but compression efficiency is reduced for intra prediction regions
Solution Approach 1:
The picture is divided into multiple slices, with intra slices containing only intra prediction blocks and inter slices containing inter prediction blocks. This segmentation allows different LCU sizes to be applied to different slice types, improving compression efficiency for intra regions while maintaining manageable coding complexity through structured organization.
Solution Approach 2:
Different LCU sizes are applied to different regions based on their prediction type: smaller LCU sizes (e.g., 64x64) are used for intra prediction regions to improve compression efficiency, while larger LCU sizes (e.g., 128x128 or 256x256) are used for inter prediction regions. This local adaptation optimizes overall compression without requiring complex per-block decisions.
2Productivity
If smaller LCU sizes are used for intra prediction regions, then compression efficiency is improved, but slice header overhead increases
Solution Approach 1:
Multiple intra prediction blocks are grouped together into intra slices, which are then assigned a single LCU size. This merging reduces the number of LCU size changes and corresponding slice headers needed, thereby reducing overhead while still achieving improved compression efficiency through the use of smaller LCU sizes in intra regions.
3Manufacturing precision
If adaptive LCU sizes are applied based on prediction type, then coding precision is improved, but processing complexity increases
Solution Approach 1:
The picture is pre-processed to identify intra prediction blocks and organize them into intra slices before LCU size assignment. This preliminary organization based on prediction type allows for systematic application of appropriate LCU sizes without requiring complex real-time decisions during coding, thereby improving coding precision while managing processing complexity.
Data Source
AI summary
This invention relates to a method of coding video data. The method comprising: determining a largest coding unit size for a picture; determining whether a largest coding unit within a picture is an intra type largest coding unit; and applying a partitioning to the largest coding unit based on whether the largest coding unit is an intra type largest coding unit.