Periocular Face Recognition Switching for Occluded Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition systems often fail when a user's face is partially occluded, leading to increased false acceptance rates or reduced usability, as maintaining high security levels becomes challenging with partial occlusions.
Innovation Solution
A method and device that utilize a neural network module trained on both full and partial face images, including periocular regions, to enhance facial recognition authentication by processing both two-dimensional and three-dimensional image data, employing innovative use of specific technical entities such as a neural network module and a secure enclave processor, which includes a secure enclave processor (SEP) for secure authentication, and a camera system with infrared illuminators to capture and process images for enhanced security and usability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full face facial recognition is used with high security restrictions, then false acceptance rate is reduced, but usability deteriorates when face is occluded
Solution Approach 1:
The system segments the facial recognition process into two distinct pathways: full-face recognition for unoccluded images and periocular recognition for occluded images. This segmentation allows the system to apply appropriate security levels to each scenario, maintaining high security for full-face recognition while enabling usability for occluded images through the alternative periocular pathway.
Solution Approach 2:
The system dynamically switches between full-face recognition and periocular recognition modes based on the detection of occlusion in the input image. This dynamic adaptation allows the system to maintain optimal security and usability by selecting the appropriate recognition mode for each specific situation, rather than using a fixed approach.
2Ease of operation
If security restrictions are relaxed to allow authentication with occluded faces, then usability is improved, but false acceptance rate increases
Solution Approach 1:
The system applies different recognition criteria and security thresholds to different facial regions based on occlusion status. For occluded images, it focuses specifically on the periocular region with appropriate security restrictions, rather than applying relaxed restrictions to the entire face. This local quality approach maintains security while enabling occlusion tolerance.
3Measurement precision
If full face recognition is used, then authentication accuracy is high, but the system fails when face is occluded
Solution Approach 1:
The system introduces an occlusion detection intermediary that analyzes the input image to determine whether occlusion is present. Based on this detection, it routes the authentication process to either full-face recognition (for high accuracy when unoccluded) or periocular recognition (for reliability when occluded). This intermediary mechanism ensures both accuracy and reliability across different scenarios.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A facial recognition authentication process operating on a device may capture an image of a user using a camera on the device. The facial recognition authentication process may include operating a full face facial recognition authentication process on the captured image or operating a partial face facial recognition authentication process on the captured image. The process may determine which process to operate (either full face or partial face) based on an assessment of an amount of occlusion in the captured image. The partial face facial recognition authentication process may be operated when there is at least some occlusion of selected features (e.g., nose and/or mouth) on the user's face in the captured image.