Video Decoding Multi-Prediction Weighted Combination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding methods face challenges in achieving high compression rates with minimal image quality degradation due to limited bandwidth, particularly in efficiently encoding and decoding motion vectors across varying block sizes and prediction modes.
Innovation Solution
The proposed solution involves determining a multi-prediction combination mode for blocks, generating prediction blocks based on multiple prediction modes, and combining them using weights, as well as adjusting reference motion vectors and split modes to improve inter prediction accuracy and coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple prediction modes are used to improve prediction accuracy, then prediction quality improves, but encoding information size increases
Solution Approach 1:
The current block is divided into multiple prediction blocks, each processed with different prediction modes. This segmentation allows the system to apply multiple prediction strategies without encoding all possible modes for the entire block, reducing the encoding overhead while maintaining prediction accuracy through selective application of different modes to different regions.
Solution Approach 2:
The prediction mode selection is made dynamically based on block characteristics such as size and content. The encoder adaptsively chooses which prediction modes to apply to which prediction blocks, rather than using a fixed set of modes for all blocks. This dynamic adaptation reduces the need to encode extensive mode information while maintaining high prediction accuracy where needed.
2Loss of information
If block size is increased to reduce encoding information, then encoding overhead decreases, but prediction accuracy deteriorates
Solution Approach 1:
Large blocks are segmented into multiple smaller prediction blocks that can be processed with different prediction modes. This allows the system to maintain low encoding overhead by working with larger block structures while achieving high prediction accuracy through fine-grained prediction at the prediction block level. Each prediction block can be optimized independently.
Solution Approach 2:
Different prediction modes are applied to different prediction blocks within a larger block structure, allowing local optimization of prediction accuracy based on local characteristics. This local quality approach ensures that each region is predicted with the most appropriate mode while maintaining overall encoding efficiency.
3Productivity
If inter prediction is used to remove temporal redundancy, then compression efficiency improves, but motion vector encoding complexity increases
Solution Approach 1:
The current block is divided into multiple prediction blocks, each with its own motion vector. This segmentation allows the encoder to use differential encoding for motion vectors, where only the differences from reference motion vectors are encoded. This reduces the complexity of motion vector encoding while maintaining the compression efficiency benefits of inter prediction.
Solution Approach 2:
The system uses feedback from reference blocks to determine prediction modes and motion vectors for current blocks. By leveraging previously decoded motion information and using it as a reference, the encoder reduces the amount of new information that needs to be encoded, thereby reducing encoding complexity while maintaining high compression efficiency.
Data Source
AI summary
A video decoding method includes determining, based on an area of a current block, whether a multi-prediction combination mode for predicting the current block by combining prediction results obtained according to a plurality of prediction modes is applied to the current block, when the multi-prediction combination mode is applied to the current block, determining the plurality of prediction modes to be applied to the current block, generating a plurality of prediction blocks of the current block, according to the plurality of prediction modes, and determining a combined prediction block of the current block, by combining the plurality of prediction blocks according to respective weights.