Cardboard Mask Attack Dataset

3k+ videos of printed cardboard mask attacks with real accessories for anti-spoofing and liveness detection

Check samples on Kaggle

Introduction

The Cardboard Mask Attack Dataset is a collection of 3,000 videos featuring a specific type of presentation attack: full-face photos printed on rigid cardboard, worn as masks, and enhanced with real physical accessories – wigs, hats, glasses. Unlike flat photo print attacks, cardboard masks with real accessories add volume, conceal mask edges, and create a more realistic appearance, making them significantly harder to detect for face anti-spoofing and liveness detection systems

Dataset summary

Parameter	Value
Total videos	3,000+
Unique participants	50+
Devices	iPhone 12, iPhone 14 Pro, Samsung S23

How It Differs From Other Print Attacks

	Photo Print	Cutout Print	Cardboard Mask (this dataset)
Material	Paper	Paper	Rigid cardboard
Coverage	Held in front of camera	Cut to face shape	Worn as a mask
Accessories	None	None	Wigs, hats, glasses
Detection difficulty	Standard	Medium	Hard - mix of real and fake signals

Source and collection methodology

All videos were recorded in-house by the Axon Labs team following a controlled recording protocol. Each mask is a full-face photo printed on rigid cardboard, combined with real physical accessories (wigs, hats, glasses). Videos include active liveness features: zoom-in/out, natural head movements, and blinking. Recordings were captured across diverse real-world backgrounds and lighting conditions

Use cases and applications

Face Anti-Spoofing & PAD. Train presentation attack detection models to identify accessory-enhanced print mask attacks, a blind spot for systems trained only on flat photo prints, replay attacks, or standard 2D masks
Liveness Detection. Improve liveness detection robustness by exposing models to attacks where real accessories (wigs, glasses, hats) add volume and realism to a printed face, challenging both texture-based and edge-detection methods
iBeta Certification Preparation. Test your liveness system against realistic 2D partial attacks before submitting to iBeta Level 1 or Level 2 certification

Dataset Features

3,000 videos with active liveness features (zoom-in/out, head movements, blinking)
Multi-device capture – iPhone 12, iPhone 14 Pro, Samsung S23 and others
Real accessories – wigs, hats, glasses layered on printed cardboard masks
Varied environments – diverse real-world backgrounds and lighting conditions

Need More Data?

This dataset is a ready-made sample. We offer custom data collection for paper mask attacks tailored to your requirements, including larger participant pools, additional devices, specific demographic distributions, and custom mask configurations

Download information

A sample version of this dataset is available on Kaggle. Leave a request for additional samples in the form below

Have a question?

Where does the data come from?

All data is collected in-house by the Axon Labs team. All information is verified by our specialists

What is the process?

Once your enquiry has been sent, we will contact you to discuss the details and complete the necessary paperwork. The timing of receiving the dataset depends on the specific request and additional requirements

Can you help us meet dataset disclosure requirements, GDPR and other regulatory controls?

Our unique selling point is to provide legally clean datasets to our customers. We obtain the consent from all the participants to use their data for AI model development. We are able to provide comprehensive reporting on the licensing, data collection and privacy compliance of our datasets

What is the price of the dataset?

The price depends on your specific requirements. Please submit a request to receive a free consultation