Selfie Id Dataset

1. About the dataset
2. Distributions
3. Content
4. Get the Dataset
5. Links

About the dataset

We introduce a large image dataset Selfie Id Dataset for training a neural network to repel various attacks on biometric access systems. The dataset consists of selfie photos and selfie videos of people. "Anti Spoofing Real Dataset" solves the tasks of training algorithms to distinguish real users from scammers. Proposed dataset allows to build identity recognition systems, which can be used to develop anti-spoofing solutions, such as countermeasures and system configurations that can help make authentication systems more secure.

The dataset consists of 44,832 videos and selfies from 37,980 unique people from 170 countries. The data for the dataset is still gathering, so the number of videos and photos is getting bigger!

The dataset includes 2 different types of images:

• Selfies - 13 selfies of a person from a mobile phone, the person is depicted alone on it, the face is clearly visible.
• Id Photos - id photos of the person from 2 different documents.

Data in the dataset

• People from 18 to 76 age old are presented in the dataset.
• For each person in the dataset age, country and gender is presented.
• The data was mostly (approximately 90%) collected indoor, however there are also selfies made outdoors.
• The lighting is artificial in 80% of cases, 5% natural daily lightning, 5% evening outdoor lighting, 10% - dark indoor lighting.
• People provided selfies where the head takes up at least 1/2 of the frame.
• Distance from the camera is approximately 20-30 centimeters.

People in the dataset

Distributions

Gender of people in the dataset

Ethnicity of people in the dataset

Age of people in the dataset

Content

The folder "Selfies ID Images dataset" includes 29 folders (18 sets from caucasians and 11 from hispanics):

• corresponding to each person in the sample
• containing of 13 selfies and 2 photos from the documents of the individual

File with the extension .csv

includes the following information for each media file:

• SetId: the identifier of the set,
• UserRace: race of the person,
• Age: the age of the person,
• Name: name of the person,
• FName: name of the file,
• URL: the link to access the media file

Get the Dataset

This is just an example of the data. If you need access to the entire dataset, contact us via sales@trainingdata.pro or leave a request on https://trainingdata.pro/data-market?utm_source=github

Links

Resource	Link
TrainingData.pro	https://trainingdata.pro/data-market?utm_source=github
TrainingData.solutions	https://trainingdata.solutions/data-market?utm_source=github
Kaggle	https://www.kaggle.com/trainingdatapro
HuggingFace	https://huggingface.co/TrainingDataPro