DATASET
Spoof attacks with printed selfies of people of different genders, ages and ethnicities. The dataset effectively solves security problems for biometric identification systems.

Anti-spoofing
Printed photo

The ability to recognize and analyze images and videos
Computer Vision
A way to identify or confirm a person's identity by his face
Face recognition
Identification and verification of the client by financial and other institutions
Know Your Cutomer
Methods of preventing deception of identification systems
Anti-spoofing
Liveness Detection
Checks whether the system interacts with a living person
10 000
spoof attacks
10-15
seconds each video
9.41
usability
Technical
specifications:
Each dataset set includes: a photo, a video of a person and a video of an attack with a printed photo.
Photo - a selfie of a person from a mobile phone, a person is depicted on it alone, the face is clearly visible.
Technical
specifications:
Each dataset set includes: a photo, a video of a person and a video of an attack with a printed photo.
Video - shot on the front camera, on it a person alternately moves his head left, right, up and down.
Technical
specifications:
Each dataset set includes: a photo, a video of a person and a video of an attack with a printed photo.
Video of the attack - a person takes a printed photo on video for 10-15 seconds
Metadata in the cvc format is prepared for each media file:
Race
Country of residence
Age
Gender
The device from which the picture was taken
Extra Services
Data collection of photos and videos by required attributes and metadata: frame rate, image resolution, demographic and geographical characteristics
Collecting photos and videos of attacks from MacBook, iPad and iPhone webcams
Collecting videos from printed photos on different paper (plain, matte, glossy) with a length of 5 seconds to 3 minutes

Didn't find the necessary information?
Leave a request for a free consultation and a test dataset!
Tell us about your project!
See other datasets