Intra Prediction Using Tree-Structured Block Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and ultra-high-definition stereo-scopic images leads to increased data volume, resulting in higher transmission and storage costs, and existing image compression technologies are not optimized for efficient encoding and decoding of such content.
Innovation Solution
A method and device for image decoding and encoding using tree structure-based block partition, subsampling, and intra prediction, which includes deriving an intra prediction mode and extended reference pixels to improve encoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If high-resolution and ultra-high-definition stereo-scopic image contents are provided, then image quality is improved, but data volume increases leading to higher transmission and storage costs
Solution Approach 1:
The current block is divided into multiple sub-blocks through tree structure-based block partition (quad-tree or penta-tree), allowing independent processing and prediction for each sub-block. This segmentation enables more efficient compression by adapting to local image characteristics while reducing overall data volume.
Solution Approach 2:
Different intra prediction modes are applied to different sub-blocks based on their local characteristics. The MPM (Most Probable Mode) list is derived specifically for each sub-block, allowing local optimization of prediction accuracy and compression efficiency rather than applying a uniform approach to the entire block.
2Loss of energy
If existing image compression technologies are used, then transmission and storage costs are reduced, but encoding efficiency for high-resolution stereo-scopic images is not optimized
Solution Approach 1:
The partition structure and prediction modes are dynamically adapted to the specific characteristics of each block and sub-block. The tree structure-based partition allows flexible division into 2, 3, or 4 sub-blocks depending on content requirements, and MPM lists are dynamically derived for each sub-block based on neighboring block information, optimizing encoding efficiency for high-resolution stereo-scopic images.
Solution Approach 2:
The invention changes the parameters of block partitioning from fixed to variable, allowing the number and size of sub-blocks to be adjusted based on image content. Additionally, prediction mode parameters are optimized for each sub-block through MPM list derivation, improving compression performance while reducing transmission costs.
3Productivity
If tree structure-based block partition is performed, then encoding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent employs nested partition structures where a current block is divided into sub-blocks, which can themselves be further partitioned. This hierarchical nesting allows efficient organization of processing tasks and enables reuse of prediction information across different levels of the hierarchy, managing complexity through structured organization.
Solution Approach 2:
MPM (Most Probable Mode) lists are derived in advance for each sub-block based on information from neighboring blocks before actual prediction is performed. This preliminary action prepares prediction candidates ahead of time, streamlining the encoding process and reducing real-time computational complexity while maintaining high encoding efficiency.
Data Source
AI summary
A video encoding/decoding method and device disclosed herein determine the current block through block division based on a tree structure and sub-sample the current block to form one or more sub-blocks from the current block, and can encode/decode the sub-blocks belonging to the current block according to a prescribed scan order.