Gaze-Based Face Authentication for Hands-Free Spoof Resistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing authentication methods face challenges in providing high accuracy and convenience while preventing spoofing and unauthorized access, particularly in situations where manual input is difficult or impractical, such as when hands are occupied, and there is a risk of third-party interception.
Innovation Solution
An authentication system that combines gaze input keyword authentication with face authentication, using a display device to present candidate responses and a camera to capture gaze direction, along with face images, to verify the target's responses and biometric data, ensuring the target is a living person by measuring gaze movement and face changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If keyword authentication is performed by inputting keywords from a keyboard or touch panel, then authentication can be performed without carrying personal belongings, but it becomes difficult to input keywords when hands are occupied and there is a high risk of third-party interception
Solution Approach 1:
The patent replaces manual keyboard or touch panel input with voice-based authentication. The system captures voice signals through a microphone, processes them to extract authentication keywords, and performs verification without requiring manual input devices. This substitution enables hands-free operation while maintaining security through voice recognition technology.
2Ease of operation
If voice authentication is used to allow hands-free operation, then convenience is improved, but the risk of unauthorized access increases because keywords can be easily heard by third parties
Solution Approach 1:
The patent introduces an intermediary processing layer between voice capture and authentication verification. The voice signal is first captured by a microphone, then processed through signal processing algorithms that extract and verify authentication keywords. This intermediary processing enables the system to authenticate users through voice while implementing security measures to prevent third-party interception of the actual authentication data.
3Ease of operation
If eye movement-based authentication is used when hands are occupied, then hands-free operation is achieved, but there is a high risk of unauthorized access because eye movements can be easily observed by surrounding third parties
Solution Approach 1:
The patent replaces visual eye movement-based authentication with voice-based authentication. Instead of using a camera to track eye movements that can be observed by third parties, the system uses a microphone to capture voice signals for authentication. This substitution maintains hands-free operation capability while eliminating the vulnerability to visual observation by surrounding third parties.
Data Source
AI summary
The output means 81 outputs a plurality of candidate responses to questions to the target on a display device.The response specification means 82 specifies a response to the question from among the candidate responses based on a target's gaze in a face image captured by an imaging device that captures a face image of the target. The response judgment means 83 makes a correctness or incorrectness judgment of the specified response. The face authentication means 84 performs target's face authentication from the face image captured by the imaging device. The biometric determination means 85 performs target's biometrics determination based on changes in the target's gaze detected from a plurality of the face image. The target authentication means 86 authenticates the target based on results of the correctness or incorrectness judgment, the biometrics determination, and the face authentication.