AxonData's picture
Update README.md
9a3f994 verified
metadata
license: cc-by-4.0
task_categories:
  - image-classification
  - image-feature-extraction
  - video-classification
  - image-segmentation
language:
  - en
tags:
  - liveness detection
  - anti-spoofing
  - biometrics
  - facial recognition
  - machine learning
  - deep learning
  - AI
  - paper mask attack
  - iBeta certification
  - PAD attack
  - security
  - ibeta
  - face recognition
  - pad
  - authentication
  - fraud
size_categories:
  - 1K<n<10K
pretty_name: Print Attack Dataset

Liveness Detection Dataset: Photo Print attack dataset (3K individuals+)

What Is a Print Attack?

A print attack is a 2D presentation attack vector against face recognition and liveness detection systems, where an attacker presents a printed photo of a real person's face to a camera to deceive biometric authentication. Print attacks are the most common and accessible spoofing technique in face anti-spoofing research and represent the entry-level attack class tested in iBeta Level 1 PAD certification under the ISO/IEC 30107-3 standard

Robust face anti-spoofing systems must detect print attacks reliably under varied conditions - different lighting, distances, capture devices, and printing qualities. NIST FATE benchmarks also include print attack scenarios with zoom-in effects to evaluate algorithm performance under camera-distance variation, which is why this dataset includes 15–20 second videos with zoom-in phases

Full version of dataset is availible for commercial usage - leave a request on our website Axon Labs to purchase the dataset 💰

Dataset Description:

  • 3,000+ Participants: Engaged in the project
  • Diverse Representation: Balanced mix of genders and ethnicities
  • 7,000+ Photo Print Attacks: Executed on the participants

Photo Print attack description:

  • Each attack comprises of 15-20 sec. video with Zoom in effects
  • High-quality photos with realistic colors
  • No visible image borders during the Zoom-in phase
  • Paper attacks conducted on flat photos with a straight view on the camera (not bent or skewed)

Academic Baseline Reference

The canonical academic benchmark for print attack anti-spoofing research is the Idiap Print-Attack Database (idiap.ch/en/scientific-research/data/printattack), published by the Idiap Research Institute as one of the foundational datasets in face anti-spoofing literature. This commercial dataset extends Idiap's research line with significantly more participants (3,000+ vs Idiap's 50), broader demographic representation, NIST-FATE-compliant zoom-in effects, and modern smartphone capture conditions, designed for production face recognition and liveness detection systems rather than research benchmarks alone

Potential Use Cases:

Liveness detection: This dataset is ideal for training and evaluating liveness detection models, enabling researchers to distinguish between selfies and photo print attacks with high accuracy

Successfull Spoofing attack on a Liveness test by Duobango

Keywords: Print photo attack dataset, Antispoofing for AI, Liveness Detection dataset for AI, Spoof Detection dataset, Facial Recognition dataset, Biometric Authentication dataset, AI Dataset, PAD Attack Dataset, Anti-Spoofing Technology, Facial Biometrics, Machine Learning Dataset, Deep Learning