Adaptive Block Super-Resolution for Complex Motion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional super-resolution techniques struggle with complex motion and occlusion in video frames, leading to poor image quality due to inaccurate subpixel registration and sensitivity to registration errors.
Innovation Solution
A block-based framework for processing super-resolution images, where the target image is partitioned into adaptively sized blocks based on image content and motion estimation accuracy, allowing for flexible enhancement techniques tailored to each block's characteristics, and incorporating quantization parameters for noise suppression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional super-resolution techniques are applied to enhance resolution, then spatial resolution is improved, but the quality deteriorates in regions with complex motion or occlusion due to inaccurate subpixel registration
Solution Approach 1:
The patent divides the target image into multiple blocks based on motion estimation accuracy. Blocks are segmented into different categories (e.g., well-registered blocks, partially registered blocks, unregistered blocks) and processed with different enhancement strategies. This segmentation allows the system to apply appropriate super-resolution techniques to each region independently, avoiding the propagation of registration errors across the entire image.
Solution Approach 2:
The patent applies different enhancement modes to different blocks based on their registration characteristics. Well-registered blocks receive standard super-resolution enhancement, while blocks with poor registration or complex motion receive alternative processing. This local differentiation ensures that each region is processed according to its specific properties, maintaining high quality throughout the image.
2Productivity
If single frame-based interpolation is used to enhance resolution, then processing speed is improved, but visual quality deteriorates due to lack of additional information
Solution Approach 1:
The patent merges information from multiple video frames to enhance the target image. Instead of using single frame-based interpolation, the system combines data from multiple frames with different temporal characteristics. This merging process leverages the temporal redundancy and consistency across frames to recover high-frequency details that are not present in any single frame, thereby improving visual quality while maintaining computational efficiency.
3Device complexity
If conventional interpolation techniques are applied to improve resolution, then processing simplicity is maintained, but resolution enhancement is insufficient
Solution Approach 1:
The patent introduces dynamic adaptation in the super-resolution process by adjusting the enhancement strategy based on block-level motion estimation results. The system dynamically selects appropriate enhancement modes for different blocks, adapting the processing complexity and methodology to the specific characteristics of each region. This dynamic approach enables effective resolution enhancement without requiring overly complex processing throughout the entire image.
Data Source
AI summary
In a method of processing a super-resolution target image from a plurality of substantially low resolution auxiliary frames, the target image is partitioned into a plurality of adaptively sized blocks, which are sized based upon registration confidence levels of the blocks obtained from information contained in the plurality of auxiliary frames. The blocks are classified into a plurality of different categories according to one or both of their respective registration confidence levels and their respective variance levels. In addition, separate enhancement modes designed to enhance the blocks are selected according to their respective classifications and applied on the blocks to enhance the target image.