Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,67 @@
|
|
1 |
-
---
|
2 |
-
license: cc-by-4.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-4.0
|
3 |
+
tags:
|
4 |
+
- "liveness detection"
|
5 |
+
- "anti-spoofing"
|
6 |
+
- "biometrics"
|
7 |
+
- "facial recognition"
|
8 |
+
- "machine learning"
|
9 |
+
- "deep learning"
|
10 |
+
- "AI"
|
11 |
+
- "3D mask attack"
|
12 |
+
- "PAD attack"
|
13 |
+
- "active liveness"
|
14 |
+
- "security"
|
15 |
+
---
|
16 |
+
# Liveness Detection Dataset: 3D Paper Mask Attacks
|
17 |
+
|
18 |
+
## The full version of the dataset is available for commercial use. Request access via our website at Axonlabs to purchase the dataset 💰
|
19 |
+
## For feedback and additional sample requests, please contact us!
|
20 |
+
|
21 |
+
## Dataset Description
|
22 |
+
|
23 |
+
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.
|
24 |
+
|
25 |
+
### Key Features
|
26 |
+
- **40+ Participants**: Engaged in the dataset creation, with a balanced representation of Caucasian, Black, and Asian ethnicities.
|
27 |
+
- **Video Capture**: Videos are captured on both **iOS and Android phones**, with **multiple frames** and **approximately 7 seconds** of video per attack.
|
28 |
+
- **Active Liveness**: Includes a **zoom-in and zoom-out phase** to simulate active liveness detection.
|
29 |
+
- **Diverse Scenarios**:
|
30 |
+
- Options to add **volume-based elements** such as scarves, glasses, and hoodies.
|
31 |
+
- Captured using both **low-end and high-end devices**.
|
32 |
+
- Includes specific **attack scenarios** and **movements**, especially useful for **active liveness testing**.
|
33 |
+
- **Specific paper types** are used for attacks, contributing to the diversity of the dataset.
|
34 |
+
|
35 |
+
### Ongoing Data Collection
|
36 |
+
- This dataset is still in the data collection phase, and we welcome feedback and requests to incorporate additional features or specific requirements.
|
37 |
+
|
38 |
+
### Potential Use Cases
|
39 |
+
This dataset is ideal for training and evaluating models for:
|
40 |
+
- **Liveness Detection**: Distinguishing between selfies and advanced spoofing attacks using 3D paper masks.
|
41 |
+
- **iBeta Liveness Testing**: Preparing models for **iBeta** liveness testing, ensuring high accuracy in differentiating real faces from spoof attacks.
|
42 |
+
- **Anti-Spoofing**: Enhancing security in biometric systems by identifying spoof attacks involving paper masks and other advanced methods.
|
43 |
+
- **Biometric Authentication**: Improving facial recognition systems' resilience to sophisticated paper-based spoofing attacks.
|
44 |
+
- **Machine Learning and Deep Learning**: Assisting researchers in developing robust liveness detection models for various testing scenarios.
|
45 |
+
|
46 |
+
### Keywords
|
47 |
+
- iBeta Certifications
|
48 |
+
- PAD Attacks
|
49 |
+
- Presentation Attack Detection
|
50 |
+
- Antispoofing
|
51 |
+
- Liveness Detection
|
52 |
+
- Spoof Detection
|
53 |
+
- Facial Recognition
|
54 |
+
- Biometric Authentication
|
55 |
+
- Security Systems
|
56 |
+
- AI Dataset
|
57 |
+
- 3D Mask Attack Dataset
|
58 |
+
- Active Liveness
|
59 |
+
- Anti-Spoofing Technology
|
60 |
+
- Facial Biometrics
|
61 |
+
- Machine Learning Dataset
|
62 |
+
- Deep Learning
|
63 |
+
|
64 |
+
## Contact and Feedback
|
65 |
+
We welcome your feedback! Feel free to reach out to us and share your experience with this dataset. If you're interested, you can also **receive additional samples for free**! 😊
|
66 |
+
|
67 |
+
Visit us at **Axonlabs** to request a full version of the dataset for commercial usage.
|