Datasets:

AxonData
/

3D_paper_mask_attack_dataset_for_Liveness

@@ -33,7 +33,7 @@ tags:
 The **3D Paper Mask Attack Dataset** focuses on **3D volume-based paper attacks**, incorporating elements such as the nose, shoulders, and forehead. These attacks are designed to be advanced and are useful for both **PAD level 1** and **level 2** liveness tests. This dataset includes videos captured using various mobile devices and incorporates active liveness detection techniques.
-### Key Features
 - **40+ Participants**: Engaged in the dataset creation, with a balanced representation of Caucasian, Black, and Asian ethnicities.
 - **Video Capture**: Videos are captured on both **iOS and Android phones**, with **multiple frames** and **approximately 7 seconds** of video per attack.
 - **Active Liveness**: Includes a **zoom-in and zoom-out phase** to simulate active liveness detection.
@@ -43,10 +43,10 @@ The **3D Paper Mask Attack Dataset** focuses on **3D volume-based paper attacks*
   - Includes specific **attack scenarios** and **movements**, especially useful for **active liveness testing**.
   - **Specific paper types** are used for attacks, contributing to the diversity of the dataset.
-### Ongoing Data Collection
 - This dataset is still in the data collection phase, and we welcome feedback and requests to incorporate additional features or specific requirements.
-### Potential Use Cases
 This dataset is ideal for training and evaluating models for:
 - **Liveness Detection**: Distinguishing between selfies and advanced spoofing attacks using 3D paper masks.
 - **iBeta Liveness Testing**: Preparing models for **iBeta** liveness testing, ensuring high accuracy in differentiating real faces from spoof attacks.
@@ -54,7 +54,7 @@ This dataset is ideal for training and evaluating models for:
 - **Biometric Authentication**: Improving facial recognition systems' resilience to sophisticated paper-based spoofing attacks.
 - **Machine Learning and Deep Learning**: Assisting researchers in developing robust liveness detection models for various testing scenarios.
-### Keywords
 - iBeta Certifications
 - PAD Attacks
 - Presentation Attack Detection

 The **3D Paper Mask Attack Dataset** focuses on **3D volume-based paper attacks**, incorporating elements such as the nose, shoulders, and forehead. These attacks are designed to be advanced and are useful for both **PAD level 1** and **level 2** liveness tests. This dataset includes videos captured using various mobile devices and incorporates active liveness detection techniques.
+## Key Features
 - **40+ Participants**: Engaged in the dataset creation, with a balanced representation of Caucasian, Black, and Asian ethnicities.
 - **Video Capture**: Videos are captured on both **iOS and Android phones**, with **multiple frames** and **approximately 7 seconds** of video per attack.
 - **Active Liveness**: Includes a **zoom-in and zoom-out phase** to simulate active liveness detection.
   - Includes specific **attack scenarios** and **movements**, especially useful for **active liveness testing**.
   - **Specific paper types** are used for attacks, contributing to the diversity of the dataset.
+## Ongoing Data Collection
 - This dataset is still in the data collection phase, and we welcome feedback and requests to incorporate additional features or specific requirements.
+## Potential Use Cases
 This dataset is ideal for training and evaluating models for:
 - **Liveness Detection**: Distinguishing between selfies and advanced spoofing attacks using 3D paper masks.
 - **iBeta Liveness Testing**: Preparing models for **iBeta** liveness testing, ensuring high accuracy in differentiating real faces from spoof attacks.
 - **Biometric Authentication**: Improving facial recognition systems' resilience to sophisticated paper-based spoofing attacks.
 - **Machine Learning and Deep Learning**: Assisting researchers in developing robust liveness detection models for various testing scenarios.
+## Keywords
 - iBeta Certifications
 - PAD Attacks
 - Presentation Attack Detection