Photo print attacks are the most common 2D presentation attack vector in biometric fraud, an attacker presents a printed photograph of a target person to a camera to impersonate that person's identity. Despite being simpler than 3D mask attacks, photo print attacks remain effective against liveness systems that rely on surface texture alone. Photo print attacks are the primary attack class tested at iBeta Level 1 PAD certification under ISO/IEC 30107-3

The dataset contains 7,000+ photo print attack videos from 3,000+ unique participants, approximately 2–3 videos per identity for within-participant variation. Each video is 10–20 seconds long, captured with front-facing (selfie) cameras on iOS and Android devices, and includes explicit zoom-in and zoom-out phases to support active liveness detection benchmarking. Prints use realistic skin-tone color reproduction and standardized flat geometry, with paper borders hidden from frame during capture to prevent trivial edge-detection bypas

The dataset targets iBeta Level 1 PAD certification - the entry tier of biometric presentation attack detection testing. iBeta Level 1 tests liveness systems against 2D attacks including photo prints, under ISO/IEC 30107-3. Photo print attacks are the most frequently attempted attack type in real-world biometric fraud, making L1 certification the foundational requirement for any production liveness system used in KYC, banking onboarding, or identity verification

This datasets target iBeta Level 1 but cover different attack classes: Photo Print Attacks use full flat printed photographs presented straight-on to the camera. Display Replay Attacks re-record videos played back on smartphone, laptop, or PC screens. For comprehensive L1 coverage, training on all of these attack classes is recommended

Photo print attacks were captured by presenting printed photographs to front-facing cameras on iOS and Android smartphones, with controlled variation in angle, distance, and lighting. Prints use high-fidelity color reproduction with realistic skin tones to eliminate obvious print artifacts. Each video follows an active liveness protocol with zoom-in and zoom-out phases, during which paper borders are hidden from frame - a deliberate anti-detection technique that prevents models from relying on visible paper edges as a shortcut for spoof detection. All captures use flat, unfolded photos presented straight-on to the camera

The dataset is built for real-world Level 1 preparation through three quality dimensions. First, scale: 7,000+ videos across 3,000+ unique identities provide the demographic diversity needed to prevent overfitting. Second, anti-detection techniques: hidden paper borders and high-fidelity print quality force models to learn genuine liveness cues rather than trivial texture shortcuts. 21% of companies that passed iBeta certification in 2025 are Axon Labs clients

Yes. All data is collected with explicit written informed consent from every participant in compliance with GDPR Article 9, which specifically governs the processing of biometric data. The dataset is licensed for commercial use in AI model training, validation, and iBeta certification preparation. Comprehensive compliance documentation, including consent provenance, collection methodology, and legal basis review, is available upon request for regulatory or audit purposes

Yes. A sample version of this dataset is available on request, you can verify attack diversity across the 3,000+ participant base, video quality, active liveness sequences, and format compatibility with your training pipeline before committing to the full dataset. Submit a request through the form on this page for sample access, typically delivered within 1–2 business days