Gesture Based Spatio-Temporal Representation Learning for Robust Fingerprint Presentation Attack Detection
Published in 2024 IEEE 18th International Conference on Automatic Face and Gesture Recognition (FG) , 2024
- Fingerprint spoof attacks are highly common and pose a significant threat to biometric security systems.
- Existing methods primarily focus on image classification, ignoring the potential benefits of temporal learning.
- The differences in the elastic properties of real versus fake fingerprints can be better detected through motion-induced gestures.
- Widely used datasets lack temporal information, prompting the creation of a new dataset to explore distortion-based spoof detection.
Recommended citation: @inproceedings{sankaran2019representation, title={GestSpoof: Gesture Based Spatio-Temporal Representation Learning For Robust Fingerprint Presentation Attack Detection}, author={Bhavin Jawade, Shreeram Subramanya, Atharv Dabhade, Srirangaraj Setlur, Venu Govindaraju}, booktitle={2024, 18th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2024)}, year={2024}, organization={IEEE} }
Download Paper