Anti-Spoofing Paper Attacks Dataset - 5,000 videos

The "Anti-Spoofing Paper Attacks Dataset - 5,000 videos" is a comprehensive collection designed for evaluating facial liveness detection systems against advanced spoofing techniques. Curated by Axon Labs, it comprises 5,000 video samples demonstrating various paper-based spoofing attacks. These attacks are crafted using high-quality printed materials such as 3D paper masks, aiming to mimic real human faces and deceive biometric authentication systems.The dataset encompasses seven distinct attack types, each representing different methods of paper-based spoofing. This diversity allows researchers to assess the robustness of liveness detection algorithms under various challenging conditions. The videos are captured in high resolution, ensuring that subtle details of the spoofing attacks are preserved for detailed analysis.This dataset serves as a valuable resource for developing and benchmarking anti-spoofing technologies, particularly in the context of facial recognition systems. By providing a large and varied set of spoofing scenarios, it enables the training and evaluation of models aimed at distinguishing between genuine and fraudulent biometric inputs.

"纸质欺骗攻击反欺诈数据集——5000段视频"（Anti-Spoofing Paper Attacks Dataset - 5,000 videos）由Axon Labs精心编纂，是一套用于评估面部活体检测系统抵御高级欺骗攻击性能的综合性数据集。该数据集包含5000段视频样本，涵盖各类基于纸质介质的欺骗攻击场景。此类攻击采用高质量印刷介质制作，例如3D纸质面具，旨在复刻真实人脸以欺骗生物特征认证系统。
该数据集涵盖7种不同的攻击类型，每一种均对应一种独特的纸质欺骗攻击手段。丰富的攻击场景多样性，可支持研究人员在多种复杂工况下评估活体检测算法的鲁棒性。所有视频均以高分辨率采集，完整保留了欺骗攻击的细微特征，便于开展精细化分析。
本数据集是开发与基准测试反欺骗技术的宝贵资源，尤其适用于面部识别系统场景。通过提供大规模且场景多元的欺骗攻击样本，该数据集可支撑针对区分真实与伪造生物特征输入的模型的训练与评估工作。