logo Idiap Research Institute        
 [BibTeX] [Marc21]
IMPROVING CROSS-DATASET PERFORMANCE OF FACE PRESENTATION ATTACK DETECTION SYSTEMS USING FACE RECOGNITION DATASETS
Type of publication: Conference paper
Citation: Mohammadi_InfoVAE_ICASSP_2020
Publication status: Accepted
Booktitle: 45th International Conference on Acoustics, Speech, and Signal Processing
Year: 2020
Publisher: IEEE
URL: https://gitlab.idiap.ch/bob/bo...
Abstract: Presentation attack detection (PAD) is now considered critically important for any face-recognition (FR) based access-control system. Current deep-learning based PAD systems show excellent performance when they are tested in intra-dataset scenarios. Under cross-dataset evaluation the performance of these PAD systems drops significantly. This lack of generalization is attributed to domain-shift. Here, we propose a novel PAD method that leverages the large variability present in FR datasets to induce invariance to factors that cause domain-shift. Evaluation of the proposed method on several datasets, including datasets collected using mobile devices, shows performance improvements in cross-dataset evaluations.
Keywords: cross-dataset evaluation, domain generalization, mobile biometrics, Presentation Attack Detection
Projects SWAN
Authors Mohammadi, Amir
Bhattacharjee, Sushil
Marcel, S├ębastien
Added by: [UNK]
Total mark: 0
Attachments
  • Mohammadi_InfoVAE_ICASSP_2020.pdf
Notes