Documents
Presentation Slides
Presentation Slides
IMPROVING CROSS-DATASET PERFORMANCE OF FACE PRESENTATION ATTACK DETECTION SYSTEMS USING FACE RECOGNITION DATASETS
- Citation Author(s):
- Submitted by:
- Amir Mohammadi
- Last updated:
- 15 May 2020 - 10:22am
- Document Type:
- Presentation Slides
- Document Year:
- 2020
- Event:
- Presenters:
- Amir Mohammadi
- Paper Code:
- 5846
- Categories:
- Log in to post comments
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.