Google, IBM & Meta Certificates – 40% Off
One plan covers every Professional Certificate on Coursera.
Unlock All Certificates
Explore multimodal video face liveness detection as a robust alternative to traditional facial authentication methods in this conference talk from DSC EUROPE 24. Efim Boieru examines the shortcomings of single-frame passive and video-based active liveness detection techniques, highlighting specific failure scenarios. Learn how combining depth estimation, motion analysis, and temporal data creates a more comprehensive approach to facial authentication security. The presentation covers the implementation of foundational models and the strategic integration of multiple modalities including RGB and depth inputs to enhance accuracy and scalability. Discover real-world applications and relevant datasets that demonstrate the practical benefits of this advanced multimodal approach to face liveness detection. This 23-minute talk was presented on November 21st at the Data Science Conference in Belgrade.
