Person Identification Using Non-Face Feature Collation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image recognition systems struggle to identify individuals in original images when only non-face features are visible, as they rely heavily on facial recognition, which is not feasible in all scenarios.
Innovation Solution
An information processing apparatus and method that acquires and processes person region feature information from both original and secondary images to detect matching non-face features, allowing for identification through collation and display of results on a single screen, even when facial recognition is impossible.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If face recognition is used for person identification, then identification accuracy is improved, but the system becomes inapplicable when facial features are not visible
Solution Approach 1:
The patent segments the person identification process into two independent parts: face recognition and non-face feature recognition. By dividing the feature extraction into facial features and non-facial features (body shape, clothing, accessories), the system can independently process each type and combine results, allowing it to function when facial features are unavailable while maintaining high accuracy through multiple feature dimensions.
Solution Approach 2:
The patent implements a universal identification system that handles multiple identification scenarios through a single framework. The system can process both face-visible and face-obscured cases using the same infrastructure, switching between or combining face recognition results and non-face feature recognition results based on input conditions, thereby achieving both high accuracy and broad adaptability.
2Speed
If only face features are used for identification, then processing speed is improved, but identification capability is lost when faces are not detectable
Solution Approach 1:
The patent performs preliminary detection to determine whether facial features are visible and detectable in the input image. Based on this preliminary assessment, the system proactively selects or activates the appropriate identification pathway (face recognition only, non-face feature recognition only, or combined approach), ensuring that identification capability is maintained while avoiding unnecessary processing steps to preserve speed.
Solution Approach 2:
The patent implements a dynamic identification system that adaptively adjusts its processing approach based on real-time conditions. The system dynamically switches between different feature extraction and comparison methods depending on whether faces are detectable, the quality of facial features, and the availability of non-face features, thereby maintaining both processing efficiency and reliable identification capability across varying scenarios.
Data Source
AI summary
An information processing apparatus (100) includes an acquisition unit (122) that acquires a first image from which person region feature information regarding a region including other than a face of a retrieval target person is extracted, a second image in which a collation result with the person region feature information indicates a match, and a facial region is detected, and result information indicating a collation result between face information stored in a storage unit and face information extracted from the facial region, and a display processing unit (130) that displays at least two of the first image, the second image, and the result information on an identical screen.