Portrait Component Segmentation for Foreground Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in accurately recognizing a foreground person in images, particularly when multiple people are present, leading to incorrect detection due to partial limbs being misidentified as foreground.
Innovation Solution
An image processing method that recognizes groups of portrait components, identifies a target region containing a human face, and separates it from other groups, blurring non-target regions to accurately distinguish foreground and background.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If regions are simply defined as portrait foreground and non-portrait background, then the image processing is simple, but the accuracy of recognizing foreground person deteriorates when multiple people are present
Solution Approach 1:
The patent segments the image into multiple groups of portrait components, where each group corresponds to one human body. This segmentation allows the system to distinguish between different people and their respective body parts, resolving the ambiguity in multi-person scenarios while maintaining manageable processing complexity.
Solution Approach 2:
The patent applies different processing qualities to different regions: the target region containing the human face is kept clear while other regions are blurred. This local differentiation ensures high recognition accuracy for the foreground person while simplifying the overall image structure.
2Quantity of substance
If partial limbs are recognized as foreground, then the detection covers more body parts, but the accuracy of foreground person recognition deteriorates due to incorrect detection
Solution Approach 1:
The patent uses visual differentiation through blurring to distinguish between target and non-target regions. The target region containing the human face maintains original clarity while other portrait components are blurred, creating a clear visual distinction that prevents misidentification of limbs as foreground.
Solution Approach 2:
The patent extracts the target region containing the human face from the rest of the portrait components. By separating the face-containing region from other body parts and applying different processing (blurring) to non-target regions, the system ensures that only the intended foreground person is accurately recognized.
3Ease of operation
If all portrait components are treated equally, then the processing is straightforward, but the ability to distinguish foreground from background deteriorates in multi-person images
Solution Approach 1:
The patent dynamically adjusts the processing applied to different portrait components based on their relationship to the target region. The system adaptively blurs non-target regions while preserving the target region, enabling reliable foreground-background distinction without requiring complex manual configuration.
Data Source
AI summary
A plurality of groups of portrait components and a region in which the plurality of groups of portrait components are located are recognized from a first image, each group of portrait components corresponding to one human body, and a target region that includes a human face is determined in the region in which the plurality of groups of portrait components are located, to blur regions other than the target region in the first target image. A target group of portrait components including the human face is recognized from the first image, so that the target region in which the target group of portrait components is located is determined as a foreground region, and limbs of other people without a human face are determined as a background region. The disclosed system and method improve the accuracy of recognizing a foreground person and reducing incorrect detection in portrait recognition.