Facial Expression Unlocking Sequence Against Photo Spoofing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing unlocking methods using human face recognition are vulnerable to security risks as they can be bypassed by using photos or face models, compromising information security.
Innovation Solution
A facial expression-based unlocking method that displays an expression unlock page with a sequence of unlock nodes, performs real-time facial expression recognition, and creates unlock status notifications based on the match between the captured facial expression and target expressions to complete unlocking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If human face recognition is used for unlocking, then user convenience is improved, but information security deteriorates due to vulnerability to photos or face models
Solution Approach 1:
The unlocking process is segmented into multiple sequential expression recognition steps rather than a single face recognition step. The system divides the authentication process into multiple nodes, each requiring a different facial expression to be recognized, thereby increasing security while maintaining user convenience.
Solution Approach 2:
The system changes the recognition parameter from static face geometry to dynamic facial expressions. By requiring recognition of specific expressions (e.g., smiling, frowning, surprised) in sequence, the system transforms the authentication mechanism from structure-based to behavior-based, preventing photo spoofing while preserving ease of use.
2Reliability
If multiple facial expressions are required for unlocking, then information security is improved, but device complexity increases
Solution Approach 1:
The system provides real-time feedback to guide users through the expression sequence automatically. The device monitors recognition progress and notifies users of their current position in the unlock node sequence, allowing the system to manage its own complexity while maintaining simple user interaction.
Solution Approach 2:
The system implements feedback mechanisms that inform users of their unlocking progress through unlock status notifications. This feedback loop manages the complexity by providing clear guidance on which expressions have been recognized and what remains, simplifying the user experience despite the multi-step process.
3Reliability
If real-time facial expression recognition is performed, then security against photo spoofing is improved, but processing time increases
Solution Approach 1:
The system performs recognition on key facial expression features rather than complete facial analysis. By focusing on specific expression characteristics needed for authentication, the system reduces processing time while maintaining security against photo spoofing through dynamic expression detection.
Data Source
AI summary
A facial expression-based unlocking method includes displaying an expression unlock page, and displaying an unlock node sequence on the expression unlock page. The method also includes creating, at a current unlock node in the unlock node sequence, an unlock status notification based on a facial expression in a facial image captured in real time, and completing unlocking based on the unlock status notification and a degree of match between the facial expression in the facial image and a corresponding target expression. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also contemplated.