Liveness Detection via Frame Signal Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current liveness detection methods are inadequate in distinguishing live users from spoofing attempts, particularly with high-definition video playback, leading to unreliable biometric authentication transactions.
Innovation Solution
A method and system that utilize a computing device to calculate parameters from face biometric data video frames, creating signals and determining a similarity score to verify user liveliness by analyzing movement and illumination changes, ensuring the score meets a threshold for authenticating live users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional liveness detection methods are used, then the system is simple to operate, but the reliability of detecting live users is insufficient
Solution Approach 1:
The patent segments the liveness detection process into multiple independent components: motion analysis (detecting natural facial movements), illumination analysis (detecting light reflection patterns), and depth analysis (using 3D structure verification). Each component processes specific features separately and contributes to the overall detection reliability, allowing the system to achieve high accuracy through multiple verification layers rather than a single complex method
Solution Approach 2:
The patent introduces intermediate processing layers between image capture and final authentication decisions. These intermediaries include motion vector calculation, illumination pattern recognition, and synthetic image generation that mediate between raw biometric data and authentication outcomes, enabling more reliable detection while maintaining manageable system complexity through modular architecture
2Measurement precision
If high definition video playback is used for spoofing, then the quality of fraudulent biometric data improves, but the difficulty of detecting spoofing attempts increases
Solution Approach 1:
The patent employs dynamic analysis that adapts to varying input qualities. The motion detection component analyzes temporal changes across video frames to detect unnatural patterns, while illumination analysis dynamically evaluates light reflection characteristics. These dynamic verification methods effectively distinguish between live users and high-definition video playback by detecting subtle temporal and optical inconsistencies that static analysis would miss
Solution Approach 2:
The patent transforms the verification approach by changing key parameters from static biometric matching to dynamic parameter analysis. Instead of relying solely on facial feature matching, the system analyzes motion vectors, illumination intensity variations, and depth parameter changes over time. This parameter transformation enables detection of spoofing attempts even when the fraudulent data quality is high, as the dynamic parameters reveal unnatural patterns inconsistent with live human physiology
Data Source
AI summary
A method for enhancing user liveness detection is provided that includes calculating, by a computing device, parameters for each frame in a video of captured face biometric data. Each parameter results from movement of at least one of the computing device and the biometric data during capture of the biometric data. The method also includes creating a signal for each parameter and calculating a similarity score. The similarity score indicates the similarity between the signals. Moreover, the method includes determining the user is live when the similarity score is at least equal to a threshold score.