--- title: "2D Masks with Eyeholes Attacks Dataset" description: "11,200+ videos 5 devices" url: "https://unidata.pro/datasets/2d-masks/" date_modified: "2025-10-08T17:01:37+03:00" language: "en-US" --- This masks with eyeholes dataset contains over 11,200 well-annotated videos of people wearing or holding 2D masks with eyeholes. Designed for facial recognition, fraud prevention, and iBeta Level 1 & 2 certification, it supports training face antispoofing models, spoofing detection algorithms, and liveness detection for real-world biometric security. ## Dataset Structure ### The Numbers Section **Numbered list:** - **Number:** 11,200+ — **Text:** videos - **Number:** 5 — **Text:** devices ### Tooltips Section **Tooltip items:** - **Name:** iBeta - **Name:** Liveness Detection - **Name:** Computer Vision - **Name:** Security - **Name:** Facial Recognition - **Name:** Anti-spoofing ### Dataset Information **Table with data:** | Characteristic | Data | | --- | --- | | Description | Videos of individuals wearing or holding 2D masks with eyeholes, | | Data types | Video | | Tasks | Liveness Detection, Computer Vision | | Total number of files | 11,200+ | | Labeling | Metadata (age, gender, ethnicity, devices) | | Gender | Male, Female | | Number of attributes | 31 | **Media Slider:** - **Video on Slayder:** - **Video on Slayder:** **Link to the sample:** [Download sample](https://drive.google.com/drive/folders/1RB-tb_a4beuyAXU9uDwFjBKePLykBcBn) ### Statistics - Charts **Charts with Titles:** - **Shortcode:** [ays_chart id='37'] — **caption above the graph:** Devices in the dataset - **Shortcode:** [ays_chart id='38'] — **caption above the graph:** Gender distribution ### Technical Specifications **Table with data:** | Characteristic | Data | | --- | --- | | Video extension | mp4, MOV | | Video Resolutions | Min = 1920х1080, Max = 3840х2160 | | Video duration | 4 second | | Number of background | 9 | | Video overlap | No more than 5% | | Devices | IOS, Android | **Source and data collection methodology:** Source and collection methodology. Data was collected via crowdsourcing platforms. ### Dataset Use Cases - Slider **Industry Cards:** - **Industry:** Biometric Security — **Title:** Enhancing Anti-Spoofing Technology — **Text:** 2D Masks with Eyeholes Attacks Dataset provides video data of masked faces for training anti-spoofing systems. It enables the development of detection algorithms that differentiate real faces from 2D mask attacks, improving face recognition, liveness detection, and biometric security in presentation attack scenarios across multiple applications. - **Industry:** Financial Services — **Title:** Preventing Fraud in Identity Verification — **Text:** Banks and fintech companies use this dataset to strengthen face authentication. By analyzing spoofing attacks and video replay, trained models can detect fake faces, reduce fraud risks, and enhance security systems, supporting safer biometric verification in digital onboarding and identity verification processes. - **Industry:** Healthcare & Access Control — **Title:** Securing Restricted Areas — **Text:** Hospitals and laboratories can leverage such datasets to protect biometric systems. The dataset provides videos of masked faces and attack scenarios, enabling the creation of anti-spoofing solutions and detection algorithms that prevent unauthorized access and improve face authentication for sensitive facilities. - **Industry:** AI Research & Certification — **Title:** Developing Deep Learning Anti-Spoofing Models — **Text:** This anti-spoofing mask dataset supports face recognition research and biometric security testing. It contains video replay and 2D mask attack data to train deep learning models capable of detecting spoofs, improving liveness detection, and preparing systems for iBeta Level 1 & 2 certification and real-world presentation attack detection. ### Fact **FAQs Heading:** FAQs **List of Questions:** - **Question:** What does 2D Masks with Eyeholes Attacks Dataset include? — **Answer:** It contains 11,200+ short video clips of individuals wearing or holding 2D masks with eyeholes. It provides demographic metadata such as age, gender, ethnicity, and device type for each recording. - **Question:** How was the dataset collected? — **Answer:** Data was gathered through crowdsourcing platforms, using both iOS and Android devices. Videos were recorded in nine different backgrounds, with resolutions ranging from 1920×1080 to 3840×2160 for realistic attack detection scenarios. - **Question:** How long are the videos? — **Answer:** Each video clip lasts around 4 seconds, making it suitable for training real-time detection algorithms in biometric authentication and liveness detection systems. - **Question:** Is it possible to request a custom anti-spoofing dataset? — **Answer:** Yes. Unidata can provide custom datasets with different attack types, mask variations, or recording conditions. This flexibility ensures your anti-spoofing systems are trained on the most relevant data. - **Question:** Do Unidata datasets follow GDPR or other data privacy regulations? — **Answer:** Yes. All Unidata datasets are fully compliant with GDPR and international data protection frameworks. This dataset was collected through legally permissible and ethical sources, containing no personally identifiable information while supporting face anti-spoofing research and biometric security development. - **Question:** How are Unidata datasets stored? — **Answer:** Unidata securely stores all datasets, including the anti-spoofing mask dataset, on AWS cloud servers. Data management aligns with ISO 27001 and ISO 27701 standards to ensure information security, availability, and privacy. - **Question:** Is this a real-world dataset or synthetic data? — **Answer:** This is a real-world dataset collected via crowdsourcing platforms. Each video showcases authentic human interactions with printed mask attacks under diverse environments, lighting conditions, and devices, making it highly suitable for training deep learning models in biometric authentication, spoofing prevention, and face recognition security research. [Full list of this site's AI-readable pages](https://unidata.pro/llms.txt)