Gaze Path Analysis for Spoofing-Resistant Identity Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image-recognition-based identity verification systems are vulnerable to spoofing, particularly using previously captured images or videos, and require complex calibration processes for accurate eye-tracking, which complicates implementation and user convenience.
Innovation Solution
A method and apparatus for active identification based on gaze path analysis that extracts and verifies the unique time-series gaze path of a user using a camera and adjustable illuminator, comparing it with registered data without requiring a display guide, and utilizing deep-learning methods to analyze 3D face shape information for authenticity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If image-recognition-based identification is used, then identification speed is improved, but security against spoofing deteriorates
Solution Approach 1:
The system performs preliminary liveness detection by analyzing gaze movement patterns before completing the identification process. The gaze following task is presented during the identification flow, and the system evaluates whether the user's gaze movements match the expected pattern of following on-screen elements, thereby detecting spoofing attempts before final authentication is granted.
Solution Approach 2:
The system provides visual feedback through on-screen elements that guide the user's gaze movement. These elements move across the screen and the system continuously monitors whether the user's gaze follows them in real-time. This feedback mechanism creates an interactive verification process that distinguishes live users from spoofing attempts while maintaining rapid identification.
2Measurement precision
If eye-tracking calibration is performed to ensure accuracy, then measurement precision is improved, but device complexity deteriorates
Solution Approach 1:
The system performs self-calibration by automatically adapting to each user's natural gaze behavior during the identification process. Instead of requiring manual calibration procedures, the system presents gaze-following tasks and learns the user's gaze patterns through multiple interactions, automatically adjusting its detection thresholds and parameters to achieve accurate measurement without user intervention.
Solution Approach 2:
The system dynamically adjusts detection parameters such as gaze velocity thresholds, acceleration limits, and tracking sensitivity based on the user's individual characteristics. By changing these parameters adaptively during and after the identification process, the system maintains high measurement precision while avoiding the need for complex pre-calibration procedures.
Data Source
AI summary
Disclosed herein are a method and apparatus for active identification based on gaze path analysis. The method may include extracting the face image of a user, extracting the gaze path of the user based on the face image, verifying the identity of the user based on the gaze path, and determining whether the face image is authentic.