Facial Expression Authentication System for Anti-Spoofing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional face recognition systems are vulnerable to photo spoofing and facemask spoofing, and require expensive dedicated equipment like 3D scanners, limiting their adoption and effectiveness in preventing identity theft.
Innovation Solution
A method and system for personal identification and authentication using facial expressions, which includes stationary face recognition, a facial expression test, continuous movement tracking, and a 3D perspective check, implemented on mobile devices without the need for dedicated hardware, to detect and prevent photo and facemask spoofing by capturing and analyzing real-time facial expressions and movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional 2D face recognition is used, then the system is simple and easy to implement, but it is vulnerable to photo spoofing attacks
Solution Approach 1:
The system transitions from static 2D image capture to dynamic video capture with multiple facial expressions and movements. The subject is required to perform dynamic actions such as changing facial expressions, blinking, and moving the head, which creates temporal dynamics that photos cannot replicate.
Solution Approach 2:
The system adds the time dimension by capturing video sequences instead of single static images. It also incorporates 3D spatial information by analyzing facial geometry from multiple angles and expressions, transforming the problem from 2D to 3D+time analysis.
2Reliability
If 3D face recognition with dedicated equipment is used, then anti-spoofing capability is improved, but the device cost and complexity increase
Solution Approach 1:
The system makes a standard camera perform multiple functions: capturing 2D images, capturing 3D depth information through multi-expression imaging, and capturing video sequences. This eliminates the need for dedicated 3D scanning equipment while achieving similar anti-spoofing capabilities.
Solution Approach 2:
Instead of using expensive dedicated 3D scanning hardware, the system creates virtual 3D models by capturing multiple 2D images from the same camera under different facial expressions and computing the 3D geometry through image processing and triangulation algorithms.
3Reliability
If multiple facial expressions and movements are required, then photo spoofing detection is improved, but the authentication time increases
Solution Approach 1:
The system structures the authentication process as a sequence of periodic actions where the subject is prompted to perform specific facial expressions and movements in a predetermined sequence, allowing the system to efficiently capture and analyze the required data within a structured time framework.
Data Source
AI summary
The present invention employs a first step of stationary face recognition, followed by a facial expression test, a continuous movement tracking test, and a 3D perspective check to identify and authenticate a subject, prevent photo spoofing and facemask spoofing, and determining whether the subject is a living person. The method requires a subject to present her face before a camera, which can be the built-in or peripheral camera of a mobile communication device. The method also requires displaying to the subject certain instructions and the real-time video feedback of the subject face on a display screen, which can be the built-in or peripheral display screen of the mobile communication device or mobile computing device. The 3D perspective check uses a single camera to take two images of the subject's face for the calculating the stereoscopic view data of the subject's face.