Subblock Difference Prediction for Intra Coding Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing encoding methods face challenges in improving intra prediction accuracy for subblocks within a coding unit (CU), leading to suboptimal compression performance due to uniform prediction modes across subblocks with varying content characteristics.
Innovation Solution
Implementing a difference prediction (DP) mode for subblocks based on preset mode indication information, allowing for finer granularity prediction and determining appropriate quantization parameters to enhance prediction accuracy and reduce bitstream overheads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a same intra prediction mode is used for all subblocks in the CU, then the prediction process is simple, but the prediction effect of some subblocks is poor
Solution Approach 1:
The patent divides the CU into multiple subblocks and applies different prediction modes to different subblocks based on their content characteristics. This segmentation allows each subblock to receive appropriate prediction treatment, improving overall prediction accuracy while maintaining processability through automated mode selection.
Solution Approach 2:
The patent applies different intra prediction modes to different subblocks within the same CU, allowing each subblock to have the prediction mode that best suits its local content characteristics. This local quality approach improves prediction accuracy for each subblock while the system manages complexity through standardized selection criteria.
2Manufacturing precision
If DP mode is applied to all subblocks, then prediction accuracy improves, but bitstream overhead increases
Solution Approach 1:
The patent applies DP mode selectively to subblocks that benefit from it, rather than uniformly across all subblocks. This local quality approach ensures high prediction accuracy for suitable subblocks while avoiding unnecessary overhead in subblocks where traditional intra prediction is sufficient.
Solution Approach 2:
The patent applies DP mode to a subset of subblocks (partial action) rather than all subblocks. This selective application achieves the necessary prediction accuracy improvement for critical subblocks while controlling bitstream overhead by avoiding excessive use of the more complex DP mode where it provides minimal benefit.
3Manufacturing precision
If prediction is performed at finer granularity, then prediction accuracy improves, but encoding complexity increases
Solution Approach 1:
The patent segments the CU into subblocks and applies different prediction modes to different subblocks. This segmentation enables finer granularity prediction while managing complexity through a systematic approach to mode selection that balances accuracy improvement with encoding feasibility.
Solution Approach 2:
The patent dynamically selects prediction modes for different subblocks based on their content characteristics, allowing the encoding system to adapt to varying complexity requirements. This dynamic approach enables fine-grained prediction where needed while maintaining simplicity where appropriate, balancing accuracy and complexity.
Data Source
AI summary
A decoding method includes determining, based on a preset mode indication information, whether to use a difference prediction mode to replace an intra prediction mode corresponding to a coding unit (CU). When the difference prediction mode is to be used, the method further includes performing difference prediction on the current subblock, to obtain a predicted value. Otherwise, the method further includes performing intra prediction based on the intra prediction mode corresponding to the CU, to obtain a predicted value.