Wrapped 3D Paper Mask Anti-Spoofing Dataset

The Wrapped 3D Paper Mask Anti-Spoofing Dataset, available on Kaggle, is designed to enhance liveness detection models by simulating advanced spoofing attacks. Specifically, it focuses on wrapped 3D paper mask attacks, a sophisticated form of 3D print attacks. The dataset comprises approximately 2,000 video clips, each featuring a subject wearing a wrapped 3D paper mask to mimic a real human face during authentication processes. These videos are intended to train and evaluate anti-spoofing systems, particularly those assessed under iBeta Level 2 protocols. The dataset is part of a broader collection that includes various spoofing techniques such as latex, silicone, and cloth-based 3D masks, offering a comprehensive resource for developing robust biometric security solutions.

可在Kaggle平台获取的包裹式3D纸膜反欺骗数据集（Wrapped 3D Paper Mask Anti-Spoofing Dataset），旨在通过模拟高级欺骗攻击优化活体检测模型。具体而言，该数据集聚焦于包裹式3D纸膜欺骗攻击——这是一类精密的3D打印类欺骗攻击形式。该数据集包含约2000段视频片段，每段视频均展示了受试者在身份验证流程中佩戴包裹式3D纸膜以模仿真实人脸的场景。此类视频可用于训练与评估反欺骗系统，尤其适用于需通过iBeta Level 2协议评估的反欺骗系统。该数据集隶属于一个涵盖多种欺骗攻击手段的更广泛数据集集合，其中包含乳胶面具、硅胶面具、织物基3D面具等攻击样本，可为开发高鲁棒性的生物识别安全解决方案提供全面的研究资源。