3D Face Authentication Using Screen-Projected Structured Light
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mainstream facial authentication systems on smartphones rely on 2D face recognition, which are vulnerable to spoofing attacks and require specialized hardware, making them resource-intensive and space-consuming, contradicting the industry's desire for efficiency and minimal resource usage.
Innovation Solution
A 3D face authentication system using a single front camera that reconstructs and recognizes 3D faces by illuminating the face with multiple light signals from the smartphone screen, employing photometric stereo and a light passcode to defend against spoofing attacks, without the need for additional hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional 2D face recognition is used, then authentication convenience is improved, but security against spoofing attacks deteriorates
Solution Approach 1:
The patent transitions from 2D face recognition to 3D face authentication by projecting structured light patterns (sequences of images with varying intensities) onto the user's face and capturing the reflected patterns. This dimensional elevation from 2D to 3D enables the system to perceive depth information and facial geometry, making it resistant to spoofing attacks while maintaining authentication convenience through smartphone camera integration.
2Reliability
If specialized hardware (infrared dot projectors and dedicated cameras) is used for 3D face authentication, then security is improved, but device complexity and resource consumption worsen
Solution Approach 1:
The patent makes the smartphone screen and camera serve multiple functions: the screen acts as both display and structured light projector, while the camera serves as both regular camera and 3D depth sensor. This multi-functionality eliminates the need for specialized infrared dot projectors and dedicated 3D cameras, reducing device complexity while maintaining security through 3D face authentication.
Solution Approach 2:
The smartphone uses its own existing components (screen and camera) to perform 3D face authentication without requiring external specialized hardware. The screen generates the structured light patterns and the camera captures the reflected patterns, making the system self-sufficient and resource-efficient while achieving secure authentication.
3Measurement precision
If specialized hardware is used for 3D authentication, then authentication accuracy is improved, but screen space and resource consumption worsen
Solution Approach 1:
The patent utilizes the smartphone's existing screen and camera for dual purposes: regular display and 3D authentication. The screen projects structured light patterns during authentication sequences, and the camera captures the reflected patterns for 3D reconstruction. This approach achieves high authentication accuracy without requiring additional specialized hardware that would consume screen space or increase resource consumption.
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
The system effectively authenticates users and defends against human, photo, and video attacks with low error rates, achieving a mean Equal Error Rate (EER) of 9.9%, 4.5%, and 0.66% respectively, while operating efficiently on commercial smartphones.
Implementation Method 1
illuminate the face with multiple light signals from the smartphone screen, employing photometric stereo
Data Source
AI summary
A non-transitory computer-readable medium encoded with a computer-readable program which, when executed by a processor, will cause a computer to execute a method of authenticating a 3-D object with a 2-D camera, the method including building a pre-determined database. The method additionally includes registering the 3-D object to a storage unit of a device comprising the 2-D camera, thereby creating a registered 3-D model of the 3-D object. Additionally, the method includes authenticating a test 3-D object by comparing the test 3-D object to the registered 3-D model.