Collected molecules will appear here. Add from search or explore.
Mobile application designed to collect behavioral biometric data (accelerometer, gyroscope, touch patterns) to support research into continuous authentication using masked one-class autoencoders.
Defensibility
stars
4
forks
1
This project is a specialized utility for academic data collection rather than a production-grade security framework. With only 4 stars and 1 fork after nearly two years, it lacks any meaningful community adoption or momentum. From a competitive standpoint, behavioral biometrics is a field dominated by heavyweight commercial players like BioCatch, LexisNexis Risk Solutions (via ThreatMetrix), and Darwinium, all of whom possess massive proprietary datasets that this project cannot replicate. Furthermore, frontier platforms (Google and Apple) are increasingly integrating behavioral signals directly into the mobile OS kernel (e.g., Android's 'Trust Agents' or Apple's 'Passkeys' evolution), making third-party collection apps redundant. The repository serves primarily as a companion to a specific academic paper ('Behavioral Biometrics-based Continuous Authentication Using A Lightweight Latent Representation Masked One-Class Autoencoder') and offers no technical moat or defensible IP beyond the research methodology itself.
TECH STACK
INTEGRATION
reference_implementation
READINESS