Image Block Prediction Using Motion Data and Template Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image coding methods are not always efficient, particularly at low and medium bitrates, as they fail to accurately predict current blocks, resulting in increased coding costs due to the use of non-pertinent reference blocks.
Innovation Solution
A method for reconstructing and coding current blocks in image sequences that involves reconstructing motion data, identifying prediction blocks using both motion data and template matching, and merging these blocks to minimize distortion, with a weighting coefficient to balance the influence of different prediction blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If template matching is used to identify reference blocks, then prediction accuracy improves, but coding complexity increases
Solution Approach 1:
The patent divides the prediction process into two independent parts: motion-compensated prediction using motion vectors and template matching using neighboring pixels. These two prediction methods operate separately and their results are combined through weighted averaging, allowing each method to be optimized independently while reducing overall coding complexity
Solution Approach 2:
The patent merges motion-compensated prediction and template matching predictions by computing a weighted average of both prediction blocks. This combination allows the system to leverage the strengths of both methods (motion vector efficiency and template accuracy) while maintaining a balanced coding complexity through a unified prediction framework
2Measurement precision
If multiple reference blocks are used for prediction, then prediction accuracy improves, but coding cost increases
Solution Approach 1:
The patent introduces a weighting parameter (alpha) that dynamically adjusts the contribution of motion-compensated prediction versus template matching prediction. By optimizing this parameter, the system achieves high prediction accuracy while controlling coding cost, as the weighted combination allows selective use of prediction methods based on their efficiency
3Productivity
If motion data is used for prediction, then coding efficiency improves, but prediction precision deteriorates at low bitrates
Solution Approach 1:
The patent uses template matching as an intermediary mechanism that leverages already-coded neighboring pixels (templates) to improve prediction precision. This intermediary approach allows the system to maintain coding efficiency while enhancing prediction accuracy, as template matching uses locally available information rather than requiring additional motion data
Data Source
AI summary
A method for decoding a stream of coded data to reconstruct a current block of a sequence of images is disclosed. The method comprises the steps of:reconstructing for the current block a first item of motion data from a stream of coded data,identifying a first prediction block with the first item of motion data,determining a second item of motion data by template matching,identifying a second prediction block with the second item of motion data,reconstructing the current block from the first and second prediction blocks.The step of determination of the second item of motion data is a function of the first distortion calculated between the first prediction block and the second prediction block.