Lip Motion Verification for Face Authentication Spoofing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face authentication technologies face increasing challenges from spoofing methods, necessitating a new living body detection method.
Innovation Solution
A living body detection support apparatus that includes a determination criteria information storage part, an instruction information output part, an image information acquisition part, a lip information acquisition part, a lip motion determination part, and a living body determination part. This apparatus instructs a subject to perform a specific lip motion, determines if the motion was performed based on acquired lip information, and concludes that the subject is a living being if the motion is correctly executed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional face authentication methods are used, then authentication can be performed in a non-contact manner, but the system becomes vulnerable to spoofing attacks using face images or videos
Solution Approach 1:
The system requires the subject to perform predetermined lip motions (opening mouth, closing mouth, protruding tongue) before authentication. These preliminary actions demonstrate that the subject is a living being capable of voluntary movement, thereby preventing spoofing attacks using static face images or videos.
Solution Approach 2:
The system captures lip motion information and compares it against the predetermined lip motions stored in the determination criteria information. This feedback mechanism verifies whether the subject's lip movements match the expected patterns, confirming liveness and preventing spoofing.
2Reliability
If lip motion determination is added to face authentication, then living body detection capability is improved, but the device complexity increases
Solution Approach 1:
The image acquisition unit serves multiple functions: it captures both the face image for authentication and the lip motion information for liveness detection. The determination criteria information storage part stores both authentication criteria and lip motion criteria, eliminating the need for separate dedicated components and reducing overall system complexity.
Data Source
AI summary
An apparatus comprising: processor and storage, wherein the storage is configured to: store determination criteria information, wherein, the determination criteria information is information in which types of lip motion and lip information for each of the types of lip motion are associated with each other, the processor is configured to: output lip motion instruction information, wherein the lip motion instruction information includes information instructing a given type of lip motion, acquire lip image information including a mouth of a subject person, acquire lip information of the subject person from the lip image information, determine whether the subject person performed the instructed lip motion, on the basis of the determination criteria information, the lip motion instruction information, and the lip information of the subject person, and determine that the subject person is a living body when it is determined that the subject person performed the instructed lip motion.