Back to datasets
Dataset assetOpen Source CommunityBiometric SecurityAttack Detection
PADISI USC Dataset
The dataset includes two subsets: PADISI‑Finger and PADISI‑Face, which are intended for research on multimodal fingerprint presentation‑attack detection and facial presentation‑attack detection, respectively.
Source
github
Created
Feb 7, 2022
Updated
May 20, 2024
Signals
193 views
Availability
Linked source ready
Overview
Dataset description and usage context
PADISI USC Dataset Overview
Dataset Composition
- PADISI‑Finger: Multimodal fingerprint presentation‑attack detection dataset. See details in
README_FINGER.md. - PADISI‑Face: Dataset for detecting and continuously learning novel facial presentation attacks, with sentence descriptions for all samples. See details in
README_FACE.md.
Sources
- PADISI‑Finger introduced in "Multi‑Modal Fingerprint Presentation Attack Detection: Evaluation On A New Dataset" (https://ieeexplore.ieee.org/document/9399674).
- PADISI‑Face introduced in "Detection and Continual Learning of Novel Face Presentation Attacks" (https://openaccess.thecvf.com/content/ICCV2021/html/Rostami_Detection_and_Continual_Learning_of_Novel_Face_Presentation_Attacks_ICCV_2021_paper.html).
- Sample descriptions for PADISI‑Face presented in "Explaining Face Presentation Attack Detection Using Natural Language" (https://doi.ieeecomputersociety.org/10.1109/FG52635.2021.9667024).
Need downstream help?
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.