---
license: cc-by-4.0
task_categories:
- image-feature-extraction
- image-classification
- video-classification
language:
- en
tags:
- liveness detection
- anti-spoofing
- biometrics
- facial recognition
- machine learning
- deep learning
- AI
- paper mask attack
- iBeta certification
- PAD attack
- security
- ibeta
- face recognition
- pad
- authentication
- fraud
---

# Black people Face Detection Dataset: 3M+ Identities

Large human faces dataset for face recognition models (10M+ images)
**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/?utm_source=hugging-face&utm_medium=cpc&utm_campaign=profile&utm_content=profile_link) to purchase the dataset 💰**


Dataset targeting 1:N and 1:1 NIST face recognition tests. Dataset contains 3M individuals, each with 3-5 images containing their faces
The dataset is “cleaned” and has low duplicate and noise level.

## Dataset Description:
- Dataset is cleaned of duplicates and non-face images
- 5M+ individuals
- Each indiviual has 3-5 face images
- Total 10M+ pictures
- Emphasis on Black peoplethnicity, while we also have other datasets targeting other ethnicities (Latin American, Asian)

## Potential Use Cases:
Facial recognition: Ideal for training models to recognize individuals across varying lighting and age conditions, crucial for 1:1 verification test NIST preparation: Perfect resource for companies aiming to excel in NIST biometric tests, offering pristine data for training and testing purposes

Keywords: NIST dataset, Facial recognition dataset, Biometric testing data, Large-scale face dataset, High-quality face images, Ethnicity-diverse face dataset, AI facial recognition training data, NIST biometric test preparation, Clean face image dataset, Identity verification dataset, facial recognition training data, 1:N facial recognition dataset