Image Interpolation Using Occlusion-Based Motion Priority
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In image processing, when a moving object's area overlaps with another object's area between temporally continuous images, multiple motion information pieces are detected at the same coordinate position, making it difficult to determine the appropriate motion information for processing, leading to potential errors in image processing, especially when depth information is unavailable.
Innovation Solution
An image processing apparatus and method that acquires motion information between images, generates occlusion information, determines the priority of motion information based on both motion and occlusion information, and performs image processing using weighted motion information for accurate interpolation and correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple motion information pieces are detected at the same coordinate position due to overlapping object areas, then motion detection coverage is improved, but image processing accuracy deteriorates due to uncertainty in selecting appropriate motion information
Solution Approach 1:
The patent introduces occlusion information as an additional parameter to distinguish between valid motion information and false motion information. By changing the parameter space from just motion information to motion information plus occlusion information, the system can accurately select appropriate motion information even when multiple pieces are detected at the same coordinate position.
Solution Approach 2:
The patent uses occlusion information as an intermediary to filter and select appropriate motion information. The occlusion information acts as a mediator between the detected motion information and the final image processing, enabling the system to identify which motion information pieces are reliable and which are false.
2Manufacturing precision
If depth information is used to select motion information, then image processing accuracy is improved, but device capability requirements worsen due to need for additional depth information acquisition
Solution Approach 1:
The patent uses occlusion information as an intermediary that can be derived from motion information itself, rather than requiring separate depth information acquisition. This mediator enables the system to select appropriate motion information without needing additional depth sensing capabilities.
Solution Approach 2:
The system uses the motion information itself to generate occlusion information, making the system self-sufficient. The motion detection process simultaneously provides both motion information and occlusion information, eliminating the need for separate depth information acquisition devices.
3Device complexity
If occlusion information is generated from motion information, then device complexity is reduced by eliminating need for separate depth sensing, but computation time increases due to additional processing steps
Solution Approach 1:
The patent merges the generation of motion information and occlusion information into a single integrated process. By combining these two functions, the system avoids the time cost of separate depth sensing while still achieving accurate motion information selection.
Solution Approach 2:
The system performs preliminary motion detection and simultaneously generates occlusion information as part of the same processing pipeline. This preliminary action ensures that both motion and occlusion information are ready for use without requiring additional sequential processing steps.
Data Source
AI summary
A motion information calculation unit acquires motion information between a plurality of target images. An occlusion information calculation unit generates occlusion information between the target images. An image interpolation processing unit determines priority of the motion information based on the motion information and the occlusion information, and performs predetermined image processing for the target images by using motion information that is weighted based on the priority.