Dark Mode
Facial Cosmetics Biometric Image Set
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This collection provides paired images of the same individuals, captured once wearing cosmetics and once without any visible beauty enhancements. The purpose of this data is to facilitate visual analysis and model training that requires clear contrast between facial states. The images are high quality, featuring a diverse range of people across various ages, ethnicities, and genders. In the images where cosmetics are applied, they are limited to specific facial areas such as the eyes, lips, or skin, allowing for precise detection and analysis.
Columns
The supporting CSV file documents the individuals and media files, containing the following fields:
- no_makeup: A link directing to the photo of the person without cosmetics.
- with_makeup: A link directing to the photo of the person wearing cosmetics.
- part: The specific body part where the cosmetics were applied (e.g., lips or skin).
- gender: The stated gender of the individual in the photograph.
- age: The age of the person at the time the photo was captured.
- country: The country of origin associated with the individual.
Distribution
This dataset consists of a structure that includes a set of paired image files and a small CSV file (
make_up.csv). The metadata file contains records for 26 unique individuals. The image folders, labelled no_makeup and with_makeup, contain the corresponding media files, which are identified using the same file names for easy pairing. All fields in the provided metadata are validated and present across all 26 records.Usage
This data product is suited for numerous applications, including:
- Developing and refining facial recognition systems.
- Creating augmented reality applications and photography filters.
- Advancing research in dermatology and skincare analysis.
- Generating personalised beauty consultations and product recommendations.
- Supporting visual analysis for social media and influencer marketing strategies.
Coverage
The demographic scope is broad, capturing individuals of various ages, with the mean age hovering around 39 years. While the initial sample indicates data for only one unique gender (Female), the overall design supports diversity in ethnicity. Geographically, the data includes five unique countries, with the majority of records originating from Russia (RU). The data structure is fixed, with an expected update frequency of 'Never'.
License
Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
Who Can Use It
- Computer Vision Developers: For training models focused on automatic facial makeup detection, image segmentation, and image-to-image translation.
- Beauty Tech Companies: To build AR tools or virtual try-on features for cosmetic products (e.g., lipstick, eye makeup).
- Academic Researchers: For studies in biometric data analysis and facial analysis, comparing before- and after-makeup shots.
Dataset Name Suggestions
- Makeup Detection Face Dataset
- Paired Cosmetics Visual Database
- Facial Cosmetics Biometric Image Set
- Before and After Makeup Facial Analysis Data
Attributes
Original Data Source: Facial Cosmetics Biometric Image Set
QUALITYLoading...