Logo des Repositoriums

Auflistung nach Schlagwort "biometric recognition"

1 - 2 von 2
  • Konferenzbeitrag
    Eyebrow Deserves Attention: Upper Periocular Biometrics
    (BIOSIG 2020 - Proceedings of the 19th International Conference of the Biometrics Special Interest Group, 2020) Nguyen, Hoang (Mark); Rattani, V; Derakhshani, Reza
    Ocular biometrics is attracting exceeding attention from research community and industry alike thanks to its accuracy, security, and ease of use in mobile devices, especially in the presence of occlusions such as masks worn during the COVID-19 pandemic. When considering the extended periocular region, eyebrows have not been getting enough attention due to their perceived low uniqueness. In this paper, we evaluate a mobile-friendly deep-learning model for eyebrow-based user authentication. Specifically, we used a fine-tuned lightCNN model for eyebrow based user authentication with promising results on a particularly challenging dataset and evaluation protocol (open-set with simulated twins). The methods achieved 0:99 AUC and 4:3% EER in VISOB dataset and 0:98 AUC and 5:6% EER on SiW datasets using closed-set and open-set analysis, respectively.
  • Konferenzbeitrag
    Eyebrow Recognition for Identifying Deepfake Videos
    (BIOSIG 2020 - Proceedings of the 19th International Conference of the Biometrics Special Interest Group, 2020) Nguyen, Hoang (Mark); Derakhshani, Reza
    Deepfake imagery that contains altered faces has become a threat to online content. Current anti-deepfake approaches usually do so by detecting image anomalies, such as visible artifacts or inconsistencies. However, with deepfake advances, these visual artifacts are becoming harder to detect. In this paper, we show that one can use biometric eyebrow matching as a tool to detect manipulated faces. Our method could provide an 0.88 AUC and 20.7% EER for deepfake detection when applied to the highest quality deepfake dataset, Celeb-DF.