Prediction Unit Partitioning for Efficient Intra Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-resolution and high-quality video data transmission and storage face increased costs due to higher data amounts, necessitating more efficient video compression techniques.
Innovation Solution
A method of decoding video by partitioning prediction units into multiple transform units and performing intra prediction on each, along with differentially setting coding units for low complexity decoding, and adapting entropy encoding modes based on coefficient properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If video compression techniques are applied to high-resolution video data, then transmission and storage costs are reduced, but encoding complexity increases
Solution Approach 1:
The prediction unit is partitioned into multiple transform units (e.g., four N×N transform units from a 2N×2N prediction unit). This segmentation allows independent processing of each transform unit, enabling more efficient compression through targeted intra prediction while managing encoding complexity through modular processing
Solution Approach 2:
Different intra prediction modes are applied to different transform units based on local characteristics. Each transform unit can be predicted using reference pixels from its specific location and direction, allowing adaptive optimization of compression efficiency for different regions while maintaining manageable encoding complexity through localized processing
2Productivity
If a prediction unit is partitioned into multiple transform units for intra prediction, then encoding efficiency is improved, but decoding complexity increases
Solution Approach 1:
The prediction unit is divided into multiple transform units that are processed independently. The decoder receives partition information indicating how the prediction unit is divided, allowing it to reconstruct each transform unit using the same intra prediction modes and reference pixels used during encoding, thereby achieving efficient decoding without excessive complexity
Solution Approach 2:
Partition information about the transform units is transmitted in advance through the bitstream. This allows the decoder to prepare the appropriate reference pixels and prediction modes before actual reconstruction, reducing decoding complexity by having the structure predetermined rather than requiring complex real-time analysis
3Quantity of substance
If entropy encoding is applied to video data, then data compression is achieved, but the number of bits required increases due to mode differentiation
Solution Approach 1:
The entropy encoding process differentiates between intra prediction modes (e.g., planar, vertical, horizontal, diagonal) and applies appropriate encoding strategies for each mode and each transform unit. This local differentiation optimizes compression by matching encoding methods to local characteristics, achieving better compression ratios without excessive bit overhead through context-adaptive encoding
Data Source
AI summary
Disclosed are a method for partitioning a block and a decoding device. A method for decoding an image comprises the steps of: partitioning one prediction unit, in which an intra prediction is performed in a predetermined intra prediction mode, into a plurality of transform units; and performing the intra prediction for each of the plurality of the transform units on the basis of the predetermined intra prediction mode. Thus, the invention increases the image encoding efficiency.