The dataset contains around 9.6k images of human faces which are both real images and those generated by AI.

The zip contains two folders: - Real Images: 5000 images of real human faces - AI-Generated Images: 4630 images of ai-generated human faces.

The Tufts Face Database is a comprehensive collection of human face images, ideal for facial recognition, biometric verification, and computer vision model training. It includes diverse data by ethnicity, age, gender, and region for robust AI development.

Images are of 102 adult faces 1350x1350 pixels in full colour. Template files mark out 189 coordinates delineating face shape, for use with Psychomorph or WebMorph.org.Self-reported age, gender and ethnicity are included in the file london_faces_info.csv. Attractiveness ratings (on a 1-7 scale from "much less attractiveness than average" to "much more attractive than average") for the neutral front faces from 2513 people (ages 17-90) are included in the file london_faces_ratings.csv.All individuals gave signed consent for their images to be "used in lab-based and web-based studies in their original or altered forms and to illustrate research (e.g., in scientific journals, news media or presentations)." Images were taken in London, UK, in April 2012.

FGnet Markup Scheme of the BioID Face Database - The BioID Face Database is being used within the FGnet project of the European Working Group on face and gesture recognition. David Cristinacce and Kola Babalola, PhD students from the department of Imaging Science and Biomedical Engineering at the University of Manchester marked up the images from the BioID Face Database. They selected several additional feature points, which are very useful for facial analysis and gesture recognition.

SCface is a database of static images of human faces. Images were taken in uncontrolled indoor environment using five video surveillance cameras of various qualities. Database contains 4160 static images (in visible and infrared spectrum) of 130 subjects. Images from different quality cameras mimic the real-world conditions and enable robust face recognition algorithms testing, emphasizing different law enforcement and surveillance use case scenarios.

The Radboud Faces Database (RaFD) is a high-quality faces database containing pictures of 67 models (including Caucasian males and females, Caucasian children, both boys and girls, and Moroccan Dutch males) displaying 8 emotional expressions (accordingly to the Facial Action Coding System): Anger, disgust, fear, happiness, sadness, surprise, contempt, and neutral.

The AR face database contains 100 classes of faces with 26 face images per class with various natural variation and occlusions.

4,866 People Large-angle and Multi-pose Faces Data. Each subject were collected 60 images under different scenes and light conditions. This data can be used for face recognition related tasks.

Fully AI generated human faces. Github page of the dataset

Children’s faces are underrepresented in face databases, and existing databases that do focus on children tend to have limitations in terms of the number of faces available and the diversity of ages and ethnicities represented. To improve the availability of children’s faces for experimental research purposes, we created a novel face database that contains 500 artificial images of children that are diverse in terms of both age (ages 3 to 10) and ethnicity (representing 15 different racial or ethnic groups). Using deep neural networks, we produced a large collection of synthetic photographs that look like naturalistic, realistic faces of children. To assess the representativeness of the dataset, adult participants (N = 585) judged the age, gender, ethnicity, and emotion of artificial faces selected from the set of 500 images. The images present a diverse array of artificial children’s faces, offering a valuable resource for research requiring children’s faces. The images and ratings are publicly available to researchers on Open Science Framework (https://osf.io/m78r4/).

This is the largest database of hyperspectral face images containing hyperspectral image cubes of 78 subjects imaged in multiple sessions. The data was captured with the CRI's VariSpec LCTF (Liquid Crystal Tunable Filter) integrated with a Photon Focus machine vision camera. There are 33 spectral bands comering the 400 - 720nm range with a 10nm step. The noise level in the dataset is relatively lower because we adapted the camera exposure time to the transmittance of the filter illumination intensity as well as CCD sensitivity in each band.

Tufts-Face-Database

Multi-modal face images (112 participants, >100,000 images in total)

7 image modalities: visible, near-infrared, thermal, computerized sketch, video, LYTRO and 3D images

Context

Tufts Face Database is the most comprehensive, large-scale (over 10,000 images, 74 females + 38 males, from more than 15 countries with an age range between 4 to 70 years old) face dataset that contains 7 image modalities: visible, near-infrared, thermal, computerized sketch, LYTRO, recorded video, and 3D images. This webpage/dataset contains the Tufts Face Database three-dimensional (3D) images. The other datasets are made available through separate links by the user.

Cross-modality face recognition is an emerging topic due to the wide-spread usage of different sensors in day-to-day life applications. The development of face recognition systems relies greatly on existing databases for evaluation and obtaining training examples for data-hungry machine learning algorithms. However, currently, there is no publicly available face database that includes more than two modalities for the same subject. In this work, we introduce the Tufts Face Database that includes images acquired in various modalities: photograph images, thermal images, near infrared images, a recorded video, a computerized facial sketch, and 3D images of each volunteer’s face. An Institutional Research Board protocol was obtained, and images were collected from students, staff, faculty, and their family members at Tufts University.

This database will be available to researchers worldwide in order to benchmark facial recognition algorithms for sketch, thermal, NIR, 3D face recognition and heterogamous face recognition.

Links to modalities of the Tufts Face Database

1. Tufts Face Database Computerized Sketches (TD_CS)

2. Tufts Face Database Thermal (TD_IR) Around+Emotion

3. Tufts Face Database Thermal Cropped (TD_IR_Cropped) Emotion only

4. Tufts Face Database Three Dimensional (3D) (TD_3D)

5. Tufts Face Database Lytro (TD_LYT) (Check Note)

6. Tufts Face Database 2D RGB Around (TD_RGB_A) (Check Note)

7. Tufts Face Database 2D RGB Emotion (TD_RGB_E) (Check Note)

8. Tufts Face Database Night Vision (NIR) (TD_NIR) (Check Note)

9. Tufts Face Database Video (TD_VIDEO) (Check Note)

10. Tufts Face Thermal2RGB Dataset

Note: Please use http instead of https. The link appears broken when https is used.

Image Acquisition

Each participant was seated in front of a blue background in close proximity to the camera. The cameras were mounted on tripods and the height of each camera was adjusted manually to correspond to the image center. The distance to the participant was strictly controlled during the acquisition process. A constant lighting condition was maintained using diffused lights.

TD_CS: Computerized facial sketches were generated using software FACES 4.0 [1], one of the most widely used software packages by law enforcement agencies, the FBI, and the US Military. The software allows researchers to choose a set of candidate facial components from the database based on their observation or memory.

TD_3D: The images were captured using a quad camera (an array of 4 cameras). Each individual was asked to look at a fixed view-point while the cameras were moved to 9 equidistant positions forming an approximate semi-circle around the individual. The 3D models were reconstructed using open-source structure-from-motion algorithms.

TD_IR_E(E stands for expression/emotion): The images were captured using a FLIR Vue Pro camera. Each participant was asked to pose with (1) a neutral expression, (2) a smile, (3) eyes closed, (4) exaggerated shocked expression, (5) sunglasses.

TD_IR_A (A stands for around): The images were captured using a FLIR Vue Pro camera. Each participant was asked to look at a fixed view-point while the cameras were moved to 9 equidistant positions forming an approximate semi-circle around the participant .

TD_RGB_E: The images were captured using a NIKON D3100 camera. Each participant was asked to pose with (1) a neutral expression, (2) a smile, (3) eyes closed, (4) exaggerated shocked expression, (5) sunglasses.

TD_RGB_A: The images were captured using a quad camera (an array of 4 visible field cameras). Each participant was asked to look at a fixed view-point while the cameras were moved to 9 equidistant positions forming an approximate semi-circle around the participant.

TD_NIR_A: The images were captured using a quad camera (an array of 4 night vision cameras). The l...

The Animal Faces Dataset is a collection of animal face images across multiple species, designed for AI, machine learning, and computer vision applications such as wildlife monitoring and image recognition.

A large-scale dataset for age estimation from facial images, including Indian Movie Face Database (IMFDB) with 19,906 labeled images and UTKFace with over 20,000 images labeled with age, gender, and ethnicity. Useful for AI, biometrics, and facial recognition research.

Origin

Originally obtained from the Yale Face Database.

Content

The database contains 165 GIF images of 15 subjects (subject01, subject02, etc.).

There are 11 images per subject, one for each of the following facial expressions or configurations: center-light, w/glasses, happy, left-light, w/no glasses, normal, right-light, sad, sleepy, surprised, and wink.

Note that the image "subject04.sad" has been corrupted and has been substituted by "subject04.normal".

Acknowledgments

It is free to use the data for research purposes. If experimental results are obtained that use images from within the database, all publications of these results should acknowledge the use of the "Yale Face Database". Without permission from Yale, images from within the database cannot be incorporated into a larger database which is then publicly distributed.

Inspiration

This data is very useful for starting experiments in face recognition.

Image Dataset of face images for compuer vision tasks

Dataset comprises 500,600+ images of individuals representing various races, genders, and ages, with each person having a single face image. It is designed for facial recognition and face detection research, supporting the development of advanced recognition systems. By leveraging this dataset, researchers and developers can enhance deep learning models, improve face verification and face identification techniques, and refine… See the full description on the dataset page: https://huggingface.co/datasets/UniDataPro/face-recognition-image-dataset.

The main purpose of building the AXIOM Face Database was to gather challenging pictures for fine-tuning object recognition algorithms for facial analysis based on state-of-the-art machine learning techniques. Additionally, the database was also developed with the aim of enriching with supplementary human-annotated pictures existing datasets that are used for training demographic face estimation algorithms. By taking into account these initial requirements, a team in the University of Siena collected pictures of faces of people who previously provided their written consent to participate in the experiment. The AXIOM Face Database includes both frontal and non-frontal images captured in multiple poses using different illumination conditions. More particularly, the database features 123 multiethnic individuals for whom 25 pictures were captured showing partially occluded faces coming from wearing glasses, sunglasses, caps, hats or scarves.

Similar face recognition has always been one of the most challenging research directions in face recognition.This project shared similar face images (SFD.zip) that we have collected so far. All images are labeld and collected from publicly available datasets such as LFW, CASIA-WebFace.We will continue to collect larger-scale data and continue to update this project.Because the data set is too large, we uploaded a compressed zip file (SFD.zip). Meanwhile here we upload a few examples for everyone to view.email: ileven@shu.edu.cn

The ORL face database includes 40 classes with 10 images per class with occlusions, scale variations and rotations.

Celeb Faces

## Overview

Celeb Faces is a dataset for object detection tasks - it contains Faces annotations for 423 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).