Mobile Device Authorization with 3D Liveness Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric security systems on mobile devices are vulnerable to presentation attacks, where unauthorized users use images or representations of verified users' biometric features to gain access, compromising device security.
Innovation Solution
A user authorization system that uses three-dimensional correspondence points and neural radiance fields (NeRF) to verify that an image is of a live person and compares the generated three-dimensional representation to verified user data, preventing unauthorized access by detecting and preventing the use of biometric feature representations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional biometric security systems are used, then user convenience is improved, but security against presentation attacks deteriorates
Solution Approach 1:
The patent transitions from two-dimensional image-based biometric verification to three-dimensional spatial verification using correspondence points. The system captures images from multiple angles and positions, extracts three-dimensional spatial information, and verifies biometric features in 3D space rather than 2D, making it resistant to flat image presentations while maintaining user convenience.
Solution Approach 2:
The patent replaces traditional mechanical/image-based verification systems with a neural network-based intelligent verification system. The neural network automatically analyzes spatial correspondence points, determines whether they represent a live person, and makes authorization decisions, substituting manual or simple image-matching verification with advanced AI-based analysis.
2Reliability
If biometric verification is implemented, then access control is improved, but vulnerability to unauthorized access using images deteriorates
Solution Approach 1:
The patent converts the harmful factor of image-based attacks into a beneficial verification mechanism. By requiring multiple images from different angles and positions, the system transforms potential attack vectors into additional verification data points. The neural network learns to distinguish between genuine multi-angle images of live persons and manipulated images, turning the attack method into a security feature.
Solution Approach 2:
The system performs preliminary verification by capturing and analyzing multiple images from different angles and positions before making an authorization decision. This preliminary action of gathering comprehensive spatial data from multiple perspectives ensures that the final verification is based on robust three-dimensional correspondence rather than single-image matching, preventing unauthorized access before it can occur.
3Reliability
If multiple verification parameters are used, then security is improved, but system complexity deteriorates
Solution Approach 1:
The patent merges multiple verification parameters (images from different angles, positions, and depths) into a unified three-dimensional correspondence verification process. Instead of treating each parameter separately, the system integrates them into a cohesive neural network model that processes all spatial information simultaneously, reducing operational complexity while maintaining high security verification standards.
Data Source
AI summary
Systems and techniques are disclosed for authorizing users based on biometric data. Images are collected from an authorized user are used to generate three-dimensional representations of the user that are stored for use in authorization operations. Images accompanying a request for authorization are first processed using liveness detection operations and then, if the images are associated with a live person, a three-dimensional representation of the person is generated using the images. If a correspondence to a verified user's three-dimensional representation is identified for the three-dimensional representation of the person requesting authorization, the request is granted.