Get started
logo-mobile
Get started
Commercial

Face Segmentation Image Dataset

This face segmentation dataset contains 87,871 images with 106 facial landmarks, detailed metadata (age, gender, ethnicity, attributes), and accurate segmentation masks, designed for facial recognition, pose estimation, expression recognition, and deep learning–based computer vision segmentation tasks.

Get in touch Download sample
  • images
    87,800+
  • landmarks
    100+
  • Data annotation
  • Anti-spoofing
  • Machine learning
  • Computer Vision
  • Facial Recognition

This face segmentation dataset contains 87,871 images with 106 facial landmarks, detailed metadata (age, gender, ethnicity, attributes), and accurate segmentation masks, designed for facial recognition, pose estimation, expression recognition, and deep learning–based computer vision segmentation tasks.

Get in touch Download sample
  • Data annotation
  • Anti-spoofing
  • Machine learning
  • Computer Vision
  • Facial Recognition
  • images
    87,800+
  • landmarks
    100+

Dataset Info

Characteristic Data
Description Video of people with annotated data for facial recognition
Data types Video
Tasks Facial recognition, Computer Vision
Number of images 87,871
Number of landmarks 106
Attributes Eye State, Orientation, Gender, Age, Cap, Race, Mouth State, Background, Glasses
Labeling Metadata (ID, ethnicity)
Gender Male, Female
Ethnicity Asian, Caucasian, African, Indian
Collecting scene Indoor, outdoor
Age Young people, middle-aged, old-aged
Download sample

Statistics

Distribution by gender
Distribution by ethnicity

Technical
Characteristics

Characteristic Data
Video extension jpg
Extension of labeling file json
Accuracy of of facial attributes and landmarks are over 97%
Source and collection methodology. Data was collected by a partner of Unidata.

Dataset Use Cases

  • Computer Vision and AI Research

    Developing Robust Segmentation Algorithms

    Face Segmentation Image Dataset supports research in image segmentation and object recognition. With manually labelled faces and precise segmentation masks, it enables developers to train and validate deep learning models for facial recognition, expression detection, and pose estimation – advancing the accuracy of modern segmentation algorithms and recognition systems.

  • Augmented and Virtual Reality

    Enhancing Realistic Face Rendering

    This facial landmarks dataset is valuable for AR/VR applications requiring accurate face parsing and 3D modeling. By providing detailed annotations of facial regions and landmarks, it helps create natural, responsive avatars. Developers use it to train models that align virtual overlays with real facial movements for immersive digital experiences.

  • Healthcare and Emotion Analysis

    Improving Diagnostic and Monitoring Tools

    Medical imaging systems use the face segmentation dataset to identify facial expressions and detect subtle physical changes. These segmentation maps and labeled faces datasets help recognize emotions, monitor fatigue, or detect early signs of conditions affecting facial muscles – enhancing both telemedicine and mental health evaluation accuracy through computer vision.

  • Security and Surveillance Systems

    Optimizing Identity and Expression Recognition

    Such datasets strengthen recognition algorithms used in access control, video surveillance, and forensic analysis. The dataset consists of human faces with detailed annotations and accurate segmentation, allowing systems to detect, analyze, and verify identities or expressions even in challenging lighting and pose variations.

FAQs

What is included in this dataset?
The dataset includes 87,871 high-resolution images of human faces with detailed segmentation masks and 106 annotated facial landmarks. Each image is enriched with metadata containing gender, age, ethnicity, and additional visual attributes such as eye state, mouth state, and presence of glasses or caps.
Can I request a sample of the dataset before purchasing or downloading it?
Yes. Unidata provides free dataset samples for testing and evaluation. You can analyze a small set of labeled face images and segmentation masks to assess data quality, annotation accuracy, and suitability for your model training objectives before purchasing the full dataset.
Is it possible to request a custom dataset?
Absolutely. Unidata offers custom dataset creation services, allowing you to specify parameters such as age range, ethnicity, image orientation, or lighting conditions. Custom facial segmentation datasets can also include additional annotations like facial landmarks or emotion labels to match specialized research or recognition goals.
How are Unidata datasets licensed?
Unidata datasets follow a dual-licensing model: free samples are available for testing and validation, while full datasets are accessible exclusively through purchase. This ensures transparency and provides researchers and developers with reliable, high-quality data for segmentation and recognition applications.
Do Unidata datasets follow GDPR or other data privacy regulations?
Yes. All Unidata datasets are curated in full compliance with GDPR and relevant international privacy laws. Every dataset is sourced through lawful means, ensuring ethical and responsible handling of human face data.
How are Unidata datasets stored?
All datasets are securely stored on AWS cloud infrastructure, guaranteeing high availability and protection. Unidata’s data storage aligns with ISO 27001 and ISO 27701 standards, ensuring that all facial segmentation data is handled in a privacy-focused and security-compliant environment.
How long does it take to receive the dataset?
After submitting a purchase request, Unidata will contact you to confirm details and complete documentation. Once the agreement is signed and payment processed, delivery of a dataset typically takes between 3 and 10 business days.
Is this a real-world dataset or synthetic data?
This dataset is composed of real-world facial images captured from diverse individuals across various environments. It reflects natural lighting conditions and authentic human expressions, making it ideal for developing robust and realistic face segmentation and recognition systems.
Still have questions about using Unidata datasets? Read our user-guides

Similar Datasets

Why Companies Trust Unidata’s Services for ML/AI

Share your project requirements, we handle the rest. Every service is tailored, executed, and compliance-ready, so you can focus on strategy and growth, not operations.

Share your project requirements, we handle the rest. Every service is tailored, executed, and compliance-ready, so you can focus on strategy and growth, not operations.

1,100+ Labelers & AI Experts

  • 920K files labeled daily across text, audio, video, and image data
  • 20+ expert-level data labeling tools for quality & speed
01

19+ Industries & Diverse Data Types

  • Finance, IT, Retail, Healthcare & more
  • Standard and specialized formats (DICOM, LiDAR)
02

Turnkey ML/AI Services

  • From data collection to validation
  • Multi-type annotation tailored to you
03

100% Legal & Secure

  • Legally sourced & stored data
  • AWS ISO 27001/27701
04

Smooth Collaboration

  • Dedicated PM & SLA guarantee
  • Europe-timezone communication
05

Need Proof?

See the results we've delivered for leading tech companies and startups

Explore our case

What our clients are saying

UniData

4 3 Reviews

PA

Paul 2025-02-21

Very Positive Experience!

The team was very responsive when requesting a specific dataset, and was able to work with us on what data we specifically needed and custom pricing for our use case. Overall a great experience, and would recommend them to others!

TH

Thorsten 2025-01-09

Very good experience

We got in touch with UniData to buy several datasets from them. Communication was very cooperative, quick, and friendly. We were able to find contract conditions that suited both parties well. I also appreciate the team's dedication to understand and address the needs of the customer. And the datasets we bought from UniData matched with our expectations.

Max Crous 2024-10-08

Data purchase

Our team got in touch with UniData for purchasing video data. The team at UniData was transparent, timely, and pleasant to communicate and negotiate with. Their samples and descriptions aligned well with the data we received. We will certainly reach out to UniData again if we're in search of 3rd party video data.

Abhijeet Zilpelwar 2025-02-26

Data is well organized and easy to…

Data is well organized and easy to consume. We could download and use it for training within few hours of receiving the data links.

Trusted by the world's biggest brands

Our Clients Love Us

Enterprise Document Automation

Document AI Lead

The dataset gave us strong value for both pilot and early-stage testing. We plan to broaden coverage as deployment scales.

Identity Verification Lab

Deputy Director

The data was good. We passed PAD level 1 from iBeta.

Ready to get started?

Tell us what you need — we’ll reply within 24h with a free estimate

    What service are you looking for? *
    What service are you looking for?
    Data Labeling
    Data Collection
    Ready-made Datasets
    Human Moderation
    Medicine
    Other (please describe below)
    What's your budget range? *
    What's your budget range?
    < $1,000
    $1,000 – $5,000
    $5,000 – $10,000
    $10,000 – $50,000
    $50,000+
    Not sure yet
    Where did you hear about Unidata? *
    Where did you hear about Unidata?
    Head of Client Success
    Andrew
    Head of Client Success

    — I'll guide you through every step, from your first
    message to full project delivery

    Thank you for your
    message

    It has been successfully sent!

    We use cookies to enhance your experience, personalize content, ads, and analyze traffic. By clicking 'Accept All', you agree to our Cookie Policy.