Dynamic Facial Feature Update for Mobile Face Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Face recognition systems on mobile devices with limited computing resources face challenges in maintaining accurate recognition performance due to changes in facial features over time, such as different angles, glasses, hairstyles, and makeup, without increasing computation demands.
Innovation Solution
A system for dynamic updating of facial features, comprising an image acquisition unit, face image standardization unit, facial feature comparison unit, and facial feature update unit, which preprocesses images, extracts and compares feature vectors, and updates prestored feature sets to maintain accurate recognition without increasing network size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the size of the face recognition network is increased to reduce the impact of face transformation on system performance, then recognition accuracy is improved, but the amount of computation increases several times
Solution Approach 1:
The system performs face image standardization (alignment, normalization, and preprocessing) before feature extraction and recognition. By preprocessing images to a unified standard format in advance, the system reduces the impact of face transformations without requiring a larger recognition network, thus improving accuracy while controlling computation amount.
2Adaptability or versatility
If facial features are updated after each successful recognition to adapt to changing face appearances, then adaptability is improved, but the reliability of updated features deteriorates due to non-representative vectors
Solution Approach 1:
The system implements a feedback mechanism where updated facial feature vectors are evaluated against existing feature sets. The update operation determines whether to replace, add, or keep original vectors based on similarity comparisons and diversity evaluation, ensuring that only representative and diverse features are stored. This feedback loop maintains reliability while improving adaptability to face changes over time.
3Measurement precision
If online face recognition with server-side comparison is used to achieve good recognition results, then recognition performance is improved, but device complexity and resource constraints are worsened
Solution Approach 1:
The system extracts and stores multiple facial feature vectors representing different states (angles, glasses, hairstyles, makeup) directly on the mobile device. By taking out the feature storage and comparison functionality from the server and implementing it locally, the system achieves good recognition performance while reducing device complexity and enabling offline operation.
Data Source
AI summary
Disclosed is a system for face recognition based on dynamic updating of facial features, comprising an image acquisition unit, a face image standardization unit, a facial feature comparison unit, and a facial feature update unit. The image acquisition unit acquires an original image which is processed by the face image standardization unit, and then the facial feature comparison unit completes extraction and comparison of a facial feature vector to determine whether the original image belongs to a user ID or a stranger, or to complete entry of the facial feature vector. Each user ID corresponds to one or more facial feature vectors. The facial feature update unit automatically updates the facial feature vector in a normal workflow to improve reliability and accuracy of face recognition. Also disclosed is a method for face recognition using the system. The disclosure has the advantages of simple deployment and simple to use, improving the accuracy of face recognition without increasing the size of a face recognition network, and may quickly and effectively adapt to changes in environment or user's appearance.