--- license: cc-by-4.0 tags: - ibeta-level-1 - ibeta-certification - face-anti-spoofing - face-liveness-detection - liveness-detection - presentation-attack-detection - pad - biometrics - face-recognition - spoofing - anti-spoofing - paper-attack - print-attack - cutout-attack - eyeholes-mask - cylinder-paper-mask - replay-attack - smartphone-replay - display-replay - iso-30107-3 - ekyc - identity-verification - biometric-authentication task_categories: - video-classification --- # iBeta Level 1 Dataset for Face Anti-Spoofing & Liveness Detection A presentation attack dataset for iBeta Level 1 PAD certification preparation, built for **face anti-spoofing**, **liveness detection**, and **biometric face recognition** systems. The dataset contains **30,000+ attack videos** covering all major Level 1 attack vectors: printed photo attacks, cutout paper masks, eyeholes masks, cylinder paper masks, photo masks worn by actors, and display-based replay attacks ![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F20109613%2Fe08288fbde67c3508247661be7933a7b%2FFrame-128-2.png?generation=1760347513273643&alt=media) ## What Is iBeta Level 1? iBeta Level 1 is the entry-level PAD certification tier from iBeta Quality Assurance, an independent NIST-NVLAP-accredited testing laboratory. Level 1 testing evaluates whether face recognition systems can defeat the most common 2D presentation attacks: paper prints, cutouts, eyeholes masks, and basic display replays. iBeta Level 1 compliance is increasingly required for biometric authentication systems deployed in eKYC, fintech onboarding, banking, and identity verification ## Dataset Description The dataset combines two attack categories with separate participant pools: - **Paper-based attacks**: 22,000+ videos from 85+ participants - **Replay attacks**: 8,000+ videos from 2,500+ selfie contributors Total: **30,000+ presentation attack videos** designed for end-to-end face anti-spoofing model training and iBeta Level 1 PAD certification preparation ## Key Features - **22,000+ Paper Mask Attacks** including print, cutout, eyeholes, cylinder-based volume effects, and 3D paper masks with structural elements (e.g., nose) - **8,000+ Replay Clips** with photos and videos replayed on smartphone (iOS/Android) and desktop monitor displays under varying brightness, distances, and angles - **Active Liveness Sequences** - zoom-in and zoom-out phases to evaluate motion-based liveness detection - **Multi-Ethnic Demographics** - balanced representation of Caucasian, Black, and Asian participants - **Multi-Device Capture** - iOS and Android phones (multiple device models) ## Full version of the dataset is available for commercial usage. Leave a request on our website [Axonlabs](https://axonlabs.pro/?utm_source=hugging-face&utm_medium=cpc&utm_campaign=profile&utm_content=profile_link) to purchase the dataset 💰 ## For feedback and additional sample requests, please [contact us](https://axonlab.ai/dataset/ibeta-level-1-dataset/?utm_source=hugging-face&utm_medium=cpc&utm_campaign=profile&utm_content=profile_link)! ## Attack Type Variations This dataset provides comprehensive coverage of iBeta Level 1 attack scenarios: - **Print and cutout paper attacks** - flat printed photos and cutouts shaped to face contours - **Cylinder paper masks** - printed photos shaped into cylinders to create volume effects - **3D paper masks** with structural facial elements (nose, etc.) - **Photo masks worn by actors** with head and eye variations - **Smartphone replay attacks** - photos/videos replayed on iOS and Android phone screens - **Monitor/laptop replay attacks** - photos/videos replayed on desktop displays ## Potential Use Cases - **iBeta Level 1 certification preparation** - train face anti-spoofing models against all L1 attack vectors before formal testing - **APCER/BPCER threshold validation** - measure Attack Presentation Classification Error Rate and Bona fide Presentation Classification Error Rate under ISO/IEC 30107-3 - **Face recognition spoof robustness testing** - evaluate production face recognition systems against paper and replay attacks - **eKYC and identity verification** - prepare biometric authentication systems for fintech and banking deployment - **Cross-attack generalization research** - analyze model performance across paper vs replay attack categories ## Academic Reference The canonical academic benchmarks for iBeta Level 1 attack vectors are the **Idiap Print-Attack Database** and **Idiap Replay-Attack Database** - foundational datasets from the Idiap Research Institute. This commercial dataset extends those research lines with significantly more participants (2,500+ vs Idiap's 50), modern smartphone capture, broader demographics, and direct iBeta certification mapping ## Related Datasets by Axon Labs - [iBeta Level 2 Certification Dataset](https://huggingface.co/datasets/AxonData/iBeta-Level-2-Certification-Dataset) - 25,000+ videos for advanced 3D mask attacks (L2) - [iBeta Level 3 Dataset](https://huggingface.co/datasets/AxonData/ibeta-level-3-dataset) - high-fidelity rubber and resin masks for the strictest L3 testing - [Photo Print Attack Dataset](https://huggingface.co/datasets/AxonData/Anti_spoofing_dataset_Print_attack) - focused 2D photo print attacks - [Display Replay Attacks Dataset](https://huggingface.co/datasets/AxonData/Display_replay_attacks) - display-based replay attacks - [Liveness Detection Dataset](https://huggingface.co/datasets/AxonData/liveness-detection-dataset) - comprehensive 11+ attack types in one dataset ## About Axon Labs Axon Labs builds biometric AI training datasets. We specialize in face liveness detection, face recognition, and voice anti-spoofing data for production identity verification, eKYC, fintech, and government applications ## Commercial Access Sample subset publicly available for evaluation. For full commercial dataset access, pricing, and licensing terms, contact [sales@axonlabs.pro](mailto:sales@axonlabs.pro) or visit [axonlab.ai](https://axonlab.ai/). ## Contact and Feedback We welcome your feedback! Feel free to reach out to us and share your experience with this dataset. If you're interested, you can also **receive additional samples for free**! 😊 Visit us at [**Axonlabs**](https://axonlab.ai/?utm_source=hugging-face&utm_medium=cpc&utm_campaign=profile&utm_content=profile_link) to request a full version of the dataset for commercial usage.