Sliced Image Inpainting via Border-Connected Cutting Paths
Solution Overview
Problem
Existing image inpainting techniques fail to effectively replace large missing image areas without visible artifacts, as they struggle to find suitable replacement data and assemble multiple smaller areas for large inpainted regions.
Innovation Solution
A method that divides the destination region into sub-pieces along cutting paths starting and ending at the border, determining these paths based on image structure, area, and length, and uses algorithms for spot detection and high-pass filtering to find suitable replacement data, with optional adaptation methods for seamless integration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional inpainting routines are used to replace large missing image areas, then the inpainting process can be performed, but visible artifacts and poor quality results occur
Solution Approach 1:
The patent divides the destination region into multiple sub-pieces along cutting paths that start and end at the border. Each sub-piece is then independently processed to find suitable replacement data. This segmentation allows the system to handle large inpainting areas by breaking them into manageable portions, avoiding the artifacts that occur when treating the entire large area as a single unit.
2Extent of automation
If automatic replacement data search is performed for large inpainted regions, then the process can be automated, but the quality of replacement data deteriorates
Solution Approach 1:
By segmenting the large destination region into smaller sub-pieces, the automatic replacement search algorithm can operate effectively on each small sub-piece independently. This maintains high replacement data quality because the search space for each sub-piece is limited and manageable, while still achieving full automation for the entire large inpainting region.
Solution Approach 2:
The patent applies different processing and search strategies to different sub-pieces based on their local characteristics. Each sub-piece can have its own optimal replacement data selected from appropriate source regions, ensuring local quality is maintained throughout the entire inpainted area rather than applying a uniform approach that would compromise overall quality.
3Device complexity
If the destination region is treated as a single unit for inpainting, then the process is simpler, but the ability to assemble multiple smaller replacement areas is lost
Solution Approach 1:
The destination region is divided into multiple sub-pieces that can be independently processed and assembled. This segmentation enables the system to assemble multiple smaller replacement areas from different source regions, providing the adaptability and versatility needed for high-quality inpainting of large and complex missing areas.
Data Source
AI summary
A method for replacing image data in a destination region that is divided into sub-pieces along one or more cutting paths, which start and end at two different points on the border, and finding replacement data for the sub-pieces. The cutting paths may be determined as a function of the type of image structured at the start and the end points. The cutting paths may also be determined as a function of the area of the sub-pieces and the lengths of the cutting paths. Optionally, the destination region may be determined by a spot detection algorithm. Further optionally, the spot detection algorithm may comprise calculation of a high pass filter, or detection of areas of luminosity and border-to-volume ratios. A method for moving an image element within an image is also provided.