Dual-Pixel Face Liveness Detection Against Spoofing Attacks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user verification systems are vulnerable to face spoofing attacks, where unauthorized users use images, videos, or masks to deceive the system into believing they are authorized users, leading to false acceptance.
Innovation Solution
A liveness detection method using a dual pixel sensor and a neural network-based model to analyze shape, texture, and context information from captured images to distinguish between live and fake faces, incorporating operations like subtraction, differential, and derivative processing to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional face verification systems are used, then user access can be granted based on basic biometric matching, but the system becomes vulnerable to face spoofing attacks using images, videos, or masks
Solution Approach 1:
The verification process is divided into multiple independent detection modules: liveness detection module that analyzes temporal variations and depth information, and object recognition module that identifies facial features. Each module operates independently and provides specific verification aspects, collectively forming a comprehensive security system that resists spoofing attacks.
Solution Approach 2:
The patent embeds multiple verification layers within a unified system architecture. The liveness detection module is nested within the verification framework, and the object recognition module is integrated at the same level. This nested structure allows the system to perform basic verification while simultaneously conducting advanced anti-spoofing checks without requiring separate standalone systems.
2Reliability
If multiple detection modules are added to improve anti-spoofing capability, then security against fake faces improves, but processing time and computational load increase
Solution Approach 1:
The system performs preliminary liveness detection to quickly identify and reject obvious spoofing attempts before proceeding to more time-consuming object recognition and verification processes. By filtering out fake faces early in the pipeline using temporal variation and depth analysis, the system reduces overall processing time while maintaining high security standards.
Solution Approach 2:
The patent applies different levels of detection intensity based on the situation. For high-risk scenarios, the system executes full verification including all liveness and object recognition modules. For low-risk cases, it can use simplified verification paths, balancing security requirements with processing time constraints.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A processor-implemented liveness detection method includes: obtaining an initial image using a dual pixel sensor; obtaining a left image and a right image from the initial image; and detecting liveness of an object included in the initial image using the left image and the right image.