Face Recognition Classification Correction via Dual Management Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for automatically grouping faces in large image datasets suffer from low accuracy in individual recognition, leading to misrecognition and inefficient user intervention to correct errors.
Innovation Solution
An apparatus and method that includes a first management unit for classifying feature information, a second management unit for managing objects, an association unit for updating classifications based on user corrections, and an input unit for inputting correction instructions, enhancing the accuracy of face grouping and recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic face grouping is performed using existing methods, then processing speed is improved, but individual recognition accuracy deteriorates
Solution Approach 1:
The patent segments the management of face data into two independent units: a first management unit that classifies feature information in units of similar features, and a second management unit that manages objects individually. This segmentation allows the system to maintain both efficient bulk processing and accurate individual recognition by handling different aspects of face data management separately.
2Device complexity
If face information is hidden after user rejection, then user interface complexity is reduced, but recognition accuracy improvement is limited
Solution Approach 1:
The patent implements a feedback mechanism where user corrections (acceptance or rejection of face groupings) are processed by the association unit to update the association between object classifications and feature information classifications. This feedback loop continuously improves recognition accuracy by learning from user interactions, while the dual-management structure prevents interface complexity from increasing.
Data Source
AI summary
An apparatus includes a first management unit configured to classify and manage feature information of a plurality of objects extracted from image data in units of similar feature information, a second management unit configured to classify and manage the plurality of objects extracted from the image data object by object, an association unit configured to associate classifications of the objects by the second management unit with classifications of the feature information by the first management unit, and an input unit configured to input a correction instruction about the classifications of the objects by the second management unit, wherein the association unit is configured to, if the correction instruction is input, update an association between the classifications of the objects and the classifications of the feature information and the second management unit is configured to correct and manage the classifications of the objects based on updating of the association.