Biometric Spoof Detection Through 3D-Semantic Information Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anti-spoofing detection systems in face recognition rely on a single type of information, such as three-dimensional data, making them prone to spoof attacks and lacking accuracy and reliability, and often require complex and expensive hardware or user collaboration.
Innovation Solution
A method that combines three-dimensional and semantic information from images to differentiate between real objects and their spoofs, using a classifier to process the combined information and output a data set indicating whether the object is real or a spoof, which can be implemented on conventional computing devices without complex hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple dedicated camera systems are used for anti-spoofing detection, then measurement precision and reliability are improved, but device complexity and cost increase
Solution Approach 1:
The patent combines multiple types of information (3D geometric data, texture information, and semantic information) into a unified analysis framework. This merging allows the system to achieve high spoof detection accuracy by processing diverse data types together rather than requiring separate dedicated camera systems for each type of information.
Solution Approach 2:
The system uses a single camera that can capture both 2D images and 3D depth information, making it a multi-functional device. This universal camera replaces the need for multiple dedicated camera systems while maintaining the ability to perform comprehensive anti-spoofing detection through combined analysis of various information types.
2Device complexity
If a single camera is used to generate a 3D model, then device complexity is reduced, but measurement precision and reliability of spoof detection deteriorate
Solution Approach 1:
The patent merges 3D geometric information from the single camera with additional texture and semantic information to compensate for the limitations of using only one camera. This combination of multiple information types enables the system to achieve high measurement precision for spoof detection despite the simplicity of the hardware.
Solution Approach 2:
The system creates a composite information structure by integrating 3D depth data, texture information, and semantic features. This composite approach is analogous to using composite materials - combining different types of data with complementary properties to achieve superior performance that none of the individual data types could provide alone.
3Device complexity
If only a single type of information is used for spoof detection, then processing complexity is reduced, but reliability of detection deteriorates
Solution Approach 1:
The patent merges multiple information types (3D geometric data, texture information, and semantic information) into a unified processing framework. This allows the system to achieve high reliability in spoof detection by analyzing the consistency and relationships between different information types, making it much harder for spoofers to fool the system with a single type of attack.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Method for differentiating a real object in an image from a spoof of the real object, the method comprising, obtaining an image comprising at least one object, wherein the object comprises at least one biometric identifier, such as a finger, a fingerprint, a face or a palm; extracting three-dimensional information and semantic information from the image, wherein the semantic information relates the at least one object in the image to the at least one biometric identifier and/or relates different objects in the image to each other; merging the extracted three-dimensional and semantic information to a combined information; processing the combined information by a classifier; outputting by the classifier a data set which indicates whether the at least one object in the image is the real object or a spoof of the real object.