Anti-spoofing 3D Face Reconstruction via Infrared Structured Light
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face recognition techniques based on 2D imaging or 3D reconstruction are vulnerable to print attacks, such as 2D photos, video playback, and 3D realistic face masks, which are not effectively distinguished from human skin.
Innovation Solution
The use of infrared structured light to reconstruct 3D face structures and detect face surface materials, allowing for the differentiation of human skin from artifacts without requiring additional expensive hardware, by projecting IR structured light patterns and analyzing the reflected signals with a modified CCD camera and image processing software.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If 2D imaging or 3D reconstruction is used for face recognition, then face recognition capability is provided, but the system becomes vulnerable to print attacks such as 2D photos, video playback, and 3D face masks
Solution Approach 1:
The patent changes the wavelength parameter of light from visible to infrared, exploiting the different absorption characteristics of human skin versus artificial materials in the infrared spectrum. This parameter change enables the system to distinguish genuine faces from spoofing attempts while maintaining 3D reconstruction capability
Solution Approach 2:
The patent introduces infrared structured light as an intermediary probe between the recognition system and the face. This intermediary carries information about both 3D structure (through geometric projection) and material composition (through differential absorption), enabling simultaneous structural and material analysis
2Reliability
If additional hardware is added to detect material composition, then anti-spoofing capability is improved, but system cost and complexity increase
Solution Approach 1:
The patent makes the existing camera system multi-functional by removing the infrared-blocking filter, allowing it to detect both visible light (for 3D structure) and infrared light (for material composition). This eliminates the need for separate infrared sensors while achieving dual functionality
Solution Approach 2:
The patent combines 3D structural information and material composition information into a single integrated system. By projecting structured infrared light patterns and capturing them with the modified camera, the system simultaneously obtains both geometric and material data without requiring separate hardware subsystems
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 in face recognition systems by effectively detecting and foiling 2D/3D print attacks and special make-ups, providing robust anti-spoofing capabilities through simultaneous 3D reconstruction and material detection.
Implementation Method 1
the light emitted from which is strongly absorbed by water molecules in human skin—but reflected by other materials that do not contain as much water as human skin
Implementation Method 2
the light emitted from which is strongly absorbed by water molecules in human skin—but reflected by other materials
Data Source
AI summary
Aspects of the present disclosure describe systems, methods, and structures for anti-spoofing 3D face reconstruction using infrared structured light that advantageously reconstructs 3D face structures for facial recognition and detect face surface material(s) such that human skin may be effectively distinguished from artifacts thereby providing additional security for facial recognition including immunity from 2D/3D print attacks including face masks and special make-ups.