Face Liveness Detection That Stops Photos, Masks and Deepfakes
Confirm that a real, live person is in front of the camera before you trust a selfie. Our liveness detection software blocks printed photos, screen replays, masks and deepfakes in a few seconds — inside your KYC flow or as a standalone API.
- 3 levels of presentation attacks
- Active liveness, Mask mode
- Built into KYC
- Web SDK + REST API
Trusted by 50+ banks, fintechs and exchanges to keep fake selfies out of onboarding.
- 3 levels of presentation attacks
- Active liveness, Mask mode
- Built into KYC
- Web SDK + REST API
Presentation Attacks Our Liveness Detection Stops
A presentation attack is any attempt to fool the camera with something that is not a live face.
Printed photos and screen replays
Paper printouts, photos with cut-out eyes, bent prints, social-media pictures and photos or videos shown on another phone or tablet — including replays with blinking and head movement.
Paper, silicone and 3D masks
Paper and cardboard masks, basic and professional silicone masks, theatrical masks, 3D-printed faces and full-size 3D heads.
Deepfake videos
Videos with an AI-generated or swapped face shown to the camera.
Simple printouts to professional masks
Our biometric liveness detection is trained on attacks from simple printouts to professional masks.
How Face Liveness Detection Works
A few seconds for the user, a clear live / not live answer for your system.
-
1. The user takes a selfie
The widget opens the front camera and guides the user to place their face in an oval. No app download.
-
2. The camera follows the face
In active mode, the user moves closer to the camera to fill a second oval; the face only moves toward the camera, so the check is quick and hard to fake with a flat image.
-
3. Models check depth, texture and motion
2D and 3D models look for signs of a real face — depth, skin texture, light reflections and natural movement — and for signs of paper, screens, masks and generated images.
-
4. You get a live / not live result
Your system receives the liveness result with a score, and the best frame is passed on to face matching against the ID photo.
Active Liveness Detection: the Mask Mode
Account opening: a live face from the first session
The user follows a simple instruction — moving closer to the camera. Active liveness adds a challenge that a static photo or a pre-recorded video cannot answer.
- Moves closer to the camera
How the Mask mode works
The user sees two ovals: the face starts in the first one and moves closer to the camera until it fills the second. The face only moves toward the camera, so the check stays quick and is hard to fake with a flat image.
Liveness Rules in Southeast Asia
Regulators across the region now expect a live face, not just a selfie. Last reviewed: October 2026.
Singapore
- Rule
- MAS guidance on digital onboarding
- Where liveness matters
- Remote onboarding where Singpass Myinfo is not used
- Rule
- BNM e-KYC Policy Document
- Where liveness matters
- Remote onboarding with biometric checks and fraud controls
- Rule
- OJK rules; Dukcapil face matching
- Where liveness matters
- Remote onboarding with face match against national data
- Rule
- BSP rules on electronic KYC
- Where liveness matters
- Remote onboarding for banks and e-money issuers
- Rule
- SBV Decision 2345 (from July 2024)
- Where liveness matters
- Face authentication against national records for transfers above VND 10 million
- Rule
- Bank of Thailand measures (from 2023)
- Where liveness matters
- Face scan for mobile banking transfers above THB 50,000
Liveness Detection Solutions by Industry
-
Banks and digital banks
Remote account opening and high-value transfers with a live face, not a photo.
-
E-wallets and payments
Stop fake and rented accounts at account upgrade, when users unlock higher limits.
-
Digital lending
Make sure the borrower is real and present before you pay out a loan.
-
Crypto exchanges
Block account takeover and multi-accounting with deepfakes at onboarding and withdrawal.
Liveness Detection Service Pricing
BASIC and ADVANCED KYC packages — from $0.44 per check
Document check plus selfie or liveness.
Liveness included
Liveness is included in our BASIC and ADVANCED KYC packages.
Standalone pricing
Need liveness on its own, for login or payments? Ask us for standalone pricing.
Why Choose NeuroVision Liveness Detection Software
Three attack levels covered
From printed photos to professional masks and 3D heads.
Active liveness, Mask mode
A guided check the user passes in seconds, and a flat image cannot.
Built into KYC
Liveness, face match and document checks in one session and one API.
Works on everyday phones
Runs in the mobile browser, with no app to install.
No face images kept beyond your settings
Original images are stored for 3 months by default, then per your plan; you control retention.
Cloud or on-premise
Deploy in our cloud or in your own data center.
Liveness Detection API and SDK
Choose Your Integration Path
Every integration includes:
- step-by-step implementation guides
- sandbox with test data
- dedicated technical support
- an account manager who speaks both business and tech
- regular check-ins
Back-office included
- scenario builder
- stop-lists
- manual review and analytics
Security and Biometric Data Protection
Face images are biometric data
Face images are biometric data. We encrypt them in transit and at rest, process them under GDPR and the privacy laws of our Southeast Asian markets, and let you set how long images are kept.
ISO/IEC 27001:2022 certified
Our information security management system is certified to ISO/IEC 27001:2022.
Information security policy
Processed under GDPR and the privacy laws of our Southeast Asian markets
Not sure which product fits? Tell us about your flow
We’ll map your onboarding and compliance steps to the right modules and set up a pilot on your real traffic.
Liveness Detection FAQ
Didn’t find your answer?
Liveness detection checks that a real, live person is in front of the camera rather than a photo, video, mask or deepfake. It is used together with face matching in remote identity verification.
Active liveness asks the user to do something the camera can verify. In our Mask mode the user moves closer to the camera until the face fills a second oval — a challenge that a printed photo or a pre-recorded video cannot answer.
Presentation attack detection (PAD) is the technical term for spotting fake faces shown to a camera — printouts, screens, masks. ISO/IEC 30107-3 is the international standard for testing it.
Liveness is part of our BASIC KYC package, from $0.44 per check including the document check. Contact us for standalone liveness pricing.
Turn on liveness in your KYC scenario and add our Web SDK or widget, or use the REST API. Most teams go live in one business day.
