Face-Controlled Liveness Verification via Sequential Target Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Face-based biometric authentication systems are vulnerable to replay attacks as they rely on static facial features, making it difficult to distinguish between live and recorded images, even when combined with other biometric modalities like voice recognition.
Innovation Solution
Implementing face-controlled liveness verification through a computing device that presents a sequential series of targets on a GUI, requiring users to hit targets using gaze or face pose, which is challenging for pre-captured or recorded media to replicate, thereby verifying the user's liveliness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If face recognition systems use static facial features for authentication, then the system is simple and fast, but it becomes vulnerable to replay attacks using photographs or videos
Solution Approach 1:
The patent applies dynamics by transitioning from static facial feature analysis to dynamic face pose tracking. The system captures multiple images at different poses and analyzes the temporal relationships and motion patterns between these images to determine liveness, making it resistant to static photograph or video attacks while maintaining authentication functionality
Solution Approach 2:
The system employs periodic action by requiring the user to perform a sequence of pose changes in a specific order (e.g., looking left, then right, then up). The authenticator captures images at each pose change and verifies the sequence, making it difficult for replay attacks to succeed since the attacker would need to precisely replicate the temporal sequence of pose changes
2Reliability
If the system combines face recognition with voice recognition to prevent replay attacks, then security improves, but the complexity and time required for authentication increases
Solution Approach 1:
The patent extracts the liveness verification function from the broader authentication system and implements it as a dedicated face pose-based module. This allows the system to maintain security against replay attacks through pose sequence analysis without necessarily integrating additional biometric modalities like voice recognition, thereby reducing overall system complexity
Solution Approach 2:
The face pose tracking system serves multiple functions: it verifies user identity, detects liveness, and provides security against replay attacks all within a single authentication flow. This multi-functionality eliminates the need for separate voice recognition or other biometric systems, reducing complexity while maintaining or improving security
3Measurement precision
If the system requires users to hit sequential targets using gaze or face pose, then liveness verification accuracy improves, but the time required for authentication increases
Solution Approach 1:
The system applies partial action by requiring only a limited sequence of pose changes (e.g., 3-5 targets) rather than exhaustive verification. This partial sequence is sufficient to achieve high liveness verification accuracy while keeping the authentication time reasonable, balancing security with user convenience
Solution Approach 2:
The authenticator provides immediate feedback by capturing images at each pose change and providing real-time guidance on whether the user has correctly reached the target pose. This feedback mechanism allows for rapid verification with minimal trial-and-error, reducing the overall authentication time while maintaining high accuracy
Data Source
AI summary
Techniques for implementing face-controlled liveness verification are provided. In one embodiment, a computing device can present, to a user, a sequential series of targets on a graphical user interface (GUI) of the computing device, where each target is a visual element designed to direct the user's attention to a location in the GUI. The computing device can further determine whether the user has successfully hit each target, where the determining comprises tracking movement of a virtual pointer controlled by the user's gaze or face pose and checking whether the user has moved the virtual pointer over each target. If the user has successfully hit each target, the computing device can conclude that the user is a live subject.