Continuous Face Authentication via Location Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing authentication technologies require users to make efforts for close-up and rely on different types of input, such as photos, which can lead to fake authentication issues.
Innovation Solution
An authentication method and apparatus that uses a deep learning model to perform continuous user authentication by tracking the change in location of a detected face area in image frames captured by a camera, without requiring user efforts for close-up and effectively distinguishing between real and fake inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing authentication technology requests user close-up and uses photo input, then authentication can be performed, but user convenience deteriorates and fake authentication risk increases
Solution Approach 1:
The system performs automatic face detection and continuous authentication without requiring user actions. The camera continuously captures image frames, the processor automatically detects face areas, tracks location changes, and performs authentication - all without user intervention, eliminating the need for close-up efforts while maintaining security
Solution Approach 2:
The system performs continuous authentication by continuously capturing image frames and continuously tracking face area location changes. This continuous monitoring enables authentication without requiring the user to perform close-up actions, as the system constantly verifies the user's presence and identity in the background
2Productivity
If existing authentication technology uses photo input, then authentication can be performed, but security against fake authentication deteriorates
Solution Approach 1:
The system transitions from static photo authentication to dynamic continuous authentication. By tracking the change in location of the face area across multiple continuous image frames, the system verifies that the input is a real, living face that can move and change position, rather than a static photo that cannot exhibit dynamic location changes
Solution Approach 2:
The system performs preliminary face detection and location tracking before final authentication. By continuously monitoring face area location changes in advance, the system can identify and reject fake inputs (photos) before they complete the authentication process, as photos cannot exhibit the expected dynamic location patterns of real faces
Data Source
AI summary
Disclosed are a method and apparatus for continuous authentication. An authentication method includes receiving image frames taken by a camera in succession, detecting a face area in the image frames, tracking a change in a location of the detected face area in the image frames, and performing continuous user authentication for the face area according to the change in the location by using the face area whose change in the location has been tracked and a deep learning model.