3D Reconstruction Spoof Detection via Single Camera
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in differentiating between live individuals and spoof representations, such as photographs, without requiring multiple cameras or complex hardware, especially in resource-constrained environments like mobile devices.
Innovation Solution
The method involves capturing a sequence of images from varying relative locations using a single camera, generating a 3D representation through photometric stereo processes like structure-from-motion or shape-from-shading, to determine if the subject is a live person or a spoof alternative, thereby preventing unauthorized access to secure systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple cameras or complex hardware are used to differentiate live individuals from spoof representations, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the liveness detection process into multiple temporal stages by capturing a sequence of images over time. Instead of relying on complex hardware, the system divides the authentication process into frame-by-frame analysis, extracting liveness indicators from temporal variations in the image sequence.
Solution Approach 2:
The patent transitions from spatial differentiation (multiple cameras) to temporal differentiation (sequence of images over time). By adding the time dimension to the detection process, the system achieves liveness detection capability without increasing spatial hardware complexity.
2Reliability
If additional cameras or special illuminators are deployed for spoof detection, then reliability is improved, but device complexity increases
Solution Approach 1:
The system uses the single camera's own capabilities to perform liveness detection by analyzing temporal variations in the image sequence. The camera serves multiple functions: capturing authentication images and providing liveness verification data, eliminating the need for separate verification hardware.
Solution Approach 2:
The single camera is designed to perform multiple functions: primary authentication imaging and secondary liveness detection. By making the camera universal, the system achieves reliable spoof detection without adding dedicated hardware components.
3Measurement precision
If a sequence of images is captured over time with varying relative locations, then liveness detection accuracy is improved, but loss of time increases
Solution Approach 1:
The system captures a sequence of images but does not require complete analysis of all frames. By performing partial analysis on key frames or extracting liveness indicators from selected portions of the sequence, the system achieves sufficient detection accuracy while reducing processing time.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances security by accurately distinguishing between live individuals and spoof representations, reducing hardware complexity and enabling effective liveness detection on devices with limited resources, such as mobile devices, without the need for additional cameras or special illuminators.
Implementation Method 1
an image acquisition device that captures a first sequence of images of a first subject. The first sequence of images is captured over a time period in which relative locations of the image acquisition device with respect to the first subject varies
Implementation Method 2
generating, by the one or more processing devices and using the first image and the second image as a stereo pair, a first three-dimensional representation of the first subject
Implementation Method 3
The photometric stereo process can include a structure-from-motion process
Implementation Method 4
The photometric stereo process can include a shape-from-shading process
Data Source
AI summary
In one aspect, a method includes receiving a sequence of images of a subject, the sequence captured by an image acquisition device over a time period in which relative locations of the image acquisition device varies with respect to the subject. A first image and a second image are selected from the first sequence of images. The first and second images represent a first relative location and a second relative location, respectively, of the image acquisition device with respect to the subject. A three-dimensional representation of the first subject is generated using the first image and the second image as a stereo pair. It is determined, based on the first three dimensional representation, that the first subject is an alternative representation of a live person, and in response, access to a secure system is prevented.