Eye Gaze Liveness Detection Against Photo and Video Spoofing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing liveness detection systems are vulnerable to photo and video spoofing attacks, compromising security in applications requiring biometric verification.
Innovation Solution
An information processing system that utilizes eye gaze detection through a processor to calculate and analyze the moving average and pattern of gaze angles, correlating them with moving objects on a screen to distinguish between live and spoofed inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional liveness detection techniques are used, then the authentication process is simple, but the system cannot distinguish between real live faces and photo or video-based spoofing attacks
Solution Approach 1:
The patent employs dynamic eye gaze analysis by tracking the movement and angle of the user's eyes across multiple frames. The system calculates gaze angles dynamically and compares them against expected patterns, making the liveness detection adaptive and difficult to spoof with static photos or videos.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring eye gaze patterns and comparing them against predefined liveness criteria. The detection process uses feedback from gaze angle calculations to determine whether the subject is real or spoofed, creating a closed-loop verification system.
2Reliability
If eye gaze analysis with multiple images and calculations is implemented, then spoofing attacks are prevented, but the processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary actions by capturing multiple images and calculating gaze angles in advance before making the final liveness determination. This allows the system to prepare verification data proactively, reducing the perceived processing time during actual authentication.
Solution Approach 2:
The system uses a limited number of key parameters (eye gaze angles from a small set of images) rather than analyzing all possible facial features. This partial action approach provides sufficient anti-spoofing capability while keeping processing requirements manageable.
Data Source
AI summary
An apparatus includes: a memory storing one or more instructions; and a processor configured to execute the one or more instructions to: obtain a plurality of images, each of the plurality of images including an eye of a subject; obtain a plurality of gaze angles, each of the plurality of gaze angles corresponding to the eye of the user; and detect liveness of the subject based on the plurality of gaze angles.