iBeta level 1 Liveness Detection dataset

The iBeta Level 1 Liveness Detection dataset, hosted on Kaggle, is designed to evaluate face anti-spoofing systems against paper-based presentation attacks. It comprises images from 85 unique individuals, each subjected to five distinct types of paper attacks. These attacks simulate real-world spoofing attempts, challenging liveness detection algorithms to differentiate between genuine and fraudulent facial presentations. The dataset serves as a benchmark for researchers and developers aiming to achieve iBeta Level 1 certification, a standard recognized by the National Institute of Standards and Technology (NIST) for assessing presentation attack detection (PAD) performance. By providing a diverse set of spoofing scenarios, it facilitates the development and evaluation of robust anti-spoofing technologies. This resource is particularly valuable for those working on enhancing biometric security systems and advancing the field of liveness detection.

Kaggle平台托管的iBeta Level 1活体检测数据集，专门用于评估面对纸质呈现攻击的人脸反欺骗系统。该数据集涵盖85名独立个体的图像数据，每位受试者需接受五种不同类型的纸质攻击测试。这些攻击场景模拟真实世界的欺骗尝试，对活体检测算法区分真实人脸与伪造人脸呈现的能力提出了严苛挑战。该数据集可作为有志于获取iBeta Level 1认证的研究人员与开发者的基准数据集——该标准由美国国家标准与技术研究院（National Institute of Standards and Technology，NIST）认可，用于评估呈现攻击检测（Presentation Attack Detection，PAD）性能。通过提供多样化的欺骗攻击场景，该数据集可助力鲁棒性反欺骗技术的研发与性能评估。该数据集资源对于致力于提升生物识别安全系统、推动活体检测领域发展的从业者而言，具有极高的应用价值。