Safeguard your business with Sumsub’s Liveness & Deepfake Detection technology, capable of catching all spoofing fraud instantly.
99%
completion rates worldwide
2-4 sec
average user check
99.98%
deepfake detection in one scan
Fabricating a deepfake and synthetic fraud has never been easier than it is today. A little know-how and 10 minutes are all it takes for a fraudster to get started. Deepfakes can be detected by spotting image anomalies or alterations during image capture. Fraudsters work hard to introduce new generative models to mask image tampering.
It doesn’t matter how hard fraudsters try Sumsub’s in-house pixel analysis and pattern recognition technology is capable of spotting fakes instantly. Up to 99.95% of deepfakes are intercepted on the first try.
Deepfakes, lookalikes, and doppelgangers*
Physical/digital images and
paper masks
Masks, wax figures, and
lifelike dolls
High-resolution videos
Sleeping/unwilling user verification
Сompanies that work with Sumsub identity verification software save costs and increase revenue, according to the “Total Economic Impact™ Of Sumsub’s Verification Platform study by Forrester Consulting. Let’s see what results you can get
240%
3.21M
4.55M
< 6
In one quick motion of the head, your users get verified instantly on their smartphone or webcam. Get pass rates up to 99% and happier clients.
Fully compatible with:
Sumsub leverages industry-standard data encryption and regular penetration tests to protect your business. That includes sophisticated deepfakes, advanced cyber threats, reverse engineering, man-in-the-middle, and replay attacks. Sumsub’s Liveness has been proven secure by iBeta tests, performed in accordance with ISO/IEC 30107-3.
Use Sumsub to integrate the services you need via Web SDK, Mobile SDK, or our RESTful API. Alternatively, you can use Unilink to onboard users via a QR code or a link to the verification form without integrating Sumsub at all. Don’t worry about interruptions with 99.996% uptime.
Compare and see how we excel beyond our competitors in key areas, as
validated by user reviews on G2
Liveness verification is an advanced detection technology designed for facial biometric checks. The neural network scans a face and creates a 3D map for analyzing the image and adapting to its changes.
The difference between the two technologies is that facial recognition checks whether a user is correct via biometric traits, and liveness detection checks whether the user is a live person.
The face and head are used in facial recognition to verify a person's identity through their facial biometric patterns and data. Facial biometrics then analyze every bit of facial data with the help of a 3D map captured via the photo verification process.
Biometric facial recognition works by using unique mathematical and dynamic patterns to scan a user's face. The system then maps these facial features from an image or video and compares this information with a database to find a match.
Liveness checks are used in KYC for onboarding new applicants and ongoing authentication. The Liveness neural network scans the user's face and creates a 3D map for analyzing the image and adapting to its changes. It takes just a second to recognize the facial features and report whether the applicant is a real person (not a deepfake, an AI-generated face image, a paper mask, screen photo, or another illicit image). Liveness also checks for duplicates and whether the user holds the account and linked documents.