Image Processing Apparatus for Precise Foreground Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques face challenges in precisely extracting target areas with complicated shapes, such as hair and fur, due to difficulties in generating accurate trimaps and alpha estimation, especially when objects have translucent portions or non-uniform features, leading to errors in alpha estimation and processing load issues.
Innovation Solution
An image processing apparatus that generates a trimap with non-uniform widths based on boundary information and distance features, allowing for precise extraction of target areas by defining foreground, background, and undefined areas, improving alpha estimation and reducing processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If object segmentation is used for binary segmentation of target area and background area, then the process is simple and fast, but precision in extraction of target areas with complicated shapes (such as hair and fur) and translucent portions is poor
Solution Approach 1:
The patent divides the image processing into two stages: first performs simple object segmentation to obtain initial target area, then performs alpha matting on the boundary region to refine the extraction. This segmentation approach allows the system to maintain fast processing while achieving high precision for complicated shapes and translucent portions.
Solution Approach 2:
The patent applies different processing methods to different regions: object segmentation is applied to the entire image for initial classification, while alpha matting is applied only to the boundary region for precise extraction. This local quality approach ensures high precision where needed without sacrificing overall processing efficiency.
2Manufacturing precision
If alpha matting is used to estimate transparency of pixels on boundary, then extraction precision is improved, but processing load increases due to generation of ternary trimap image and iterative alpha estimation
Solution Approach 1:
The patent segments the processing region by identifying a boundary region between target and non-target areas, then applies alpha matting only to this boundary region rather than the entire image. This reduces the processing load significantly while maintaining extraction precision where it matters most.
Solution Approach 2:
The patent performs alpha matting partially - only on the boundary region rather than the entire image - which reduces processing load while still achieving the necessary precision for complicated shapes and translucent portions that typically appear at boundaries.
3Ease of manufacture
If uniform width is used for undefined area in trimap, then generation is simple, but precision in alpha estimation deteriorates for objects with non-uniform features
Solution Approach 1:
The patent sets the width of the undefined area locally based on the distance to the boundary for each pixel, rather than using a uniform width. This allows the trimap generation to adapt to local features of the object, improving alpha estimation precision for objects with non-uniform features while maintaining reasonable simplicity in the generation process.
Data Source
AI summary
An image processing apparatus including: an acquisition unit configured to acquire a boundary between a target area and a non-target area from an image; a setting unit configured to set an undefined area on the periphery of the boundary in a width based on feature quantities of peripheral pixels of the boundary; and a generating unit configured to define an area excluding the undefined area from the target area as the foreground area, define an area excluding the undefined area from the non-target area as the background area, and generate area information that specifies the foreground area, the background area, and the undefined area.