AxonData/2D_print_attack_dataset
收藏资源简介:
--- license: cc-by-4.0 task_categories: - image-classification - image-feature-extraction - video-classification - image-segmentation language: - en tags: - biology size_categories: - 1K<n<10K --- ## Photo Print Attacks Dataset: 1K Individuals Face Anti-Spoofing Liveness dataset videos with Zoom in effect, High-Res Print ## Share with us your feedback and recieve additional samples for free!😊 ## Full version of dataset is availible for commercial usage - leave a request on our website [Axon Labs](https://axonlabs.pro/) to purchase the dataset 💰 ## Photo Print attack dataset (1K individuals+) for Presentation Attack Detection level 1 (PAD) This dataset focuses on photo print attacks which is used by both iBeta and NIST FATE to assess liveness detection algorithms. This dataset is tailored for training AI models to identify photo print attacks on individuals. Print photo attacks include Zoom effects as mandated by NIST FATE for improved AI training. ## Dataset Description: - 1,000+ Participants: Engaged in the project - Diverse Representation: Balanced mix of genders and ethnicities - 1,000+ Photo Print Attacks: Executed on the participants ## Photo Print attack description: - Each attack comprises of 15-20 sec. video with Zoom in effects - High-quality photos with realistic colors - No visible image borders during the Zoom-in phase - Paper attacks conducted on flat photos with a straight view on the camera (not bent or skewed) ## Potential Use Cases: Liveness detection: This dataset is ideal for training and evaluating liveness detection models, enabling researchers to distinguish between selfies and photo print attacks with high accuracy Keywords: Print photo attack dataset, Antispoofing for AI, Liveness Detection dataset for AI, Spoof Detection dataset, Facial Recognition dataset, Biometric Authentication dataset, AI Dataset, PAD Attack Dataset, Anti-Spoofing Technology, Facial Biometrics, Machine Learning Dataset, Deep Learning
许可协议:CC BY 4.0(知识共享署名4.0国际许可协议) 任务类别: - 图像分类 - 图像特征提取 - 视频分类 - 图像分割 语言:英语 标签:生物学 样本规模:1000至10000条样本 ## 照片打印攻击数据集:1000名个体 该数据集为面部防欺骗活体检测(Face Anti-Spoofing Liveness)视频数据集,包含带有变焦效果的高清打印照片攻击样本。 欢迎向我们反馈您的使用体验,可免费获取额外样本😊 数据集完整版支持商业使用——请前往我们的官网[Axon Labs](https://axonlabs.pro/)提交申请以购买该数据集💰 ## 面向一级呈现攻击检测(Presentation Attack Detection, PAD)的照片打印攻击数据集(1000名个体+) 本数据集被iBeta与美国国家标准与技术研究院(National Institute of Standards and Technology, NIST)的FATE(面部防欺骗测试与评估,Face Anti-Spoofing Testing and Evaluation)项目用于评估活体检测算法,专为训练用于识别个体面部照片打印攻击的人工智能模型打造。根据NIST FATE的要求,本次数据集包含变焦效果的打印照片攻击样本,以优化人工智能模型的训练效果。 ## 数据集描述: - 参与人数:1000余名受试者参与本项目 - 多样性覆盖:性别与种族分布均衡 - 攻击样本数量:1000余组照片打印攻击样本,均施加于受试者面部 ## 照片打印攻击说明: - 每组攻击样本为一段15至20秒的视频,包含变焦效果 - 照片画质高清,色彩还原真实自然 - 变焦过程中无可见图像边框 - 攻击载体为平面照片,拍摄视角与相机呈正对状态,无弯曲或歪斜情况 ## 潜在应用场景: 活体检测:本数据集非常适合用于训练和评估活体检测模型,可帮助研究人员高精度区分自拍照片与照片打印攻击样本。 ## 关键词: 照片打印攻击数据集、人工智能防欺骗、人工智能活体检测数据集、欺骗检测数据集、人脸识别数据集、生物特征认证数据集、人工智能数据集、呈现攻击检测(PAD)攻击数据集、防欺骗技术、面部生物特征、机器学习数据集、深度学习
数据集概述
数据集名称
Photo Print Attacks Dataset: 1K Individuals
数据集类型
- 图像分类
- 图像特征提取
- 视频分类
- 图像分割
语言
- 英语
标签
- 生物学
数据规模
- 1K<n<10K
数据集描述
- 参与者数量:1,000+
- 多样性:性别和种族平衡
- 攻击类型:1,000+ 照片打印攻击
攻击描述
- 视频时长:每段攻击视频包含15-20秒的缩放效果
- 照片质量:高分辨率,真实色彩
- 边界:缩放过程中无可见图像边界
- 攻击方式:平面照片,直接对准摄像头(无弯曲或倾斜)
潜在应用场景
- 活体检测:适用于训练和评估活体检测模型,区分自拍照和照片打印攻击
关键词
- 打印照片攻击数据集
- AI防欺骗
- AI活体检测数据集
- 欺骗检测数据集
- 面部识别数据集
- 生物识别认证数据集
- AI数据集
- PAD攻击数据集
- 防欺骗技术
- 面部生物识别
- 机器学习数据集
- 深度学习




