Trusted in production by
FaceNova powers face liveness for apps, driver safety monitoring, automated attendance, school bus check-in, and KYC verification — all running on-device, all in real time.
Trusted in production by
Whether you're building an app, running a fleet, managing a school, or verifying identities — we have an AI face solution ready for you.
Block fake logins with passive liveness detection. Works silently from a single frame. Flutter, Android, iOS in 3 lines.
Camera watches eyes, head and hands in real time. Drowsy? Distracted? On phone? Instant alert before accidents happen.
Walk in front of the camera — AI recognises faces and marks attendance automatically. No badges, no cards, no friction.
Students board the bus — face recognition marks them present and notifies parents instantly. No cards, no roll calls.
Verify government IDs, passports and licenses. Match the face on the document with the live user. OCR extracts all data.
Built for developers. Trusted by enterprises. Works offline.
No blink, no turn, no smile required. Our neural network detects spoofing attacks silently from a single frame — photos, videos, 3D masks all blocked.
128-dimension FaceNet embeddings for high-accuracy 1:N matching. Compare faces in under 1ms in pure Dart.
All inference runs locally on the device. No images sent to any server. Full GDPR compliance out of the box.
Per-app-package-name licensing via Firebase Remote Config. Models are encrypted at rest, decrypted only for licensed apps.
Liveness check completes in under 300ms. TFLite models optimized for mobile — minimal battery and memory impact.
await FlutterFaceNova.initialize(licenseKey: licenseKey); final result = await FlutterFaceNova.startLiveness(context); // result.isReal, result.score, result.imageBytes
The Flutter SDK liveness pipeline — 4 steps, all on-device, under 300ms.
The SDK launches a full-screen liveness camera. No extra permissions needed beyond the standard camera permission.
Our TFLite model analyzes the face in real-time. The user just looks at the camera — no actions needed.
A 512-byte FaceNet embedding is extracted — a unique mathematical fingerprint of the face.
Compare against enrolled templates using cosine similarity. Score ≥ 0.65 is a strong match.
Add the Flutter package, call two lines, done. Works on Android and iOS.
# Add to your pubspec.yaml dependencies: flutter_face_nova: git: url: https://github.com/facenova/flutter_face_nova.git permission_handler: ^11.4.0
void main() async { WidgetsFlutterBinding.ensureInitialized(); // Verify license and prepare models — call once at startup await FlutterFaceNova.initialize(licenseKey: 'YOUR_LICENSE_KEY'); await Permission.camera.request(); runApp(MyApp()); }
final LivenessResult? result = await FlutterFaceNova.startLiveness(context); if (result == null) return; // user cancelled if (!result.isReal) { print('Spoof detected. Score: ${(result.score * 100).toStringAsFixed(1)}%'); return; } // result.score → liveness confidence (0.0–1.0) // result.imageBytes → JPEG of the verified face print('Real face confirmed: ${(result.score * 100).toStringAsFixed(1)}%');
// Liveness + match in a single camera session final FaceMatchResult? match = await FlutterFaceNova.startLivenessAndMatch( context, enrolledFaces: myEnrolledFaces, matchThreshold: 75.0, ); if (match != null && match.isMatch) { print('Identity verified: ${match.matchedFace?.name} (${match.score.toStringAsFixed(1)}%)'); } // Or compare two embeddings directly (no camera, pure Dart) final double score = FlutterFaceNova.compareFaceMetadata(meta1.metadata, meta2.metadata); // Returns 0–100. ≥ 75 is a confident match.
Employee check-in and check-out with face verification. Eliminate buddy punching.
Verify users during app onboarding. Combine with ID documents for full KYC.
Replace PINs and passwords with face-based login. Fast and seamless UX.
Verify student identity before and during online exams. Prevent impersonation.
Patient identity verification before accessing medical records or services.
Grant or deny physical and digital access based on face verification.
MediaPipe-powered facial landmark analysis detects drowsiness, distraction, yawning and phone use — millisecond alerts before accidents happen. Runs on any camera, any vehicle, no GPU required.
Register employees or students once — identify and mark attendance instantly via webcam, CCTV feed or group photos. No badges, no cards, no friction.
Continuous webcam feed identifies registered faces in real-time, logging each match with timestamp and confidence score.
Upload a single group photo — AURA detects every face, matches against the database and marks attendance for the whole room at once.
Process recorded CCTV footage — the system scans every second, extracts faces, identifies known individuals and generates a full timeline report.
Upload a single query photo to instantly search your entire registered database — find who is in an image in under 300 ms regardless of database size.
Students board the bus — FaceNova's camera recognises each face, marks them present, and sends an instant notification to parents. Works offline, no internet required at the bus.
Scan any government-issued document — passport, driver's license, national ID — extract all data with OCR, then match the face on the document with the live user's face. Full KYC in under 5 seconds.
FaceNova's web platform lets you test face recognition, attendance, driver monitoring and more — directly in your browser. No install, no signup required.
Pay once and own it forever, or start with $1/month. No per-scan fees, no hidden costs.
Enterprise or custom package count? Drop us an email.
Yes, completely. No internet connection is needed at all. The license is validated entirely on-device using the license key we provide you — no network call is made. Just add the key to your code and it works offline from day one.
Never. All AI inference runs locally using on-device TFLite models. No face images, liveness scores, or biometric embeddings ever leave the device. GDPR compliance is baked in by design.
Flutter (Android + iOS) is fully available right now via the flutter_face_nova package. Native Android SDK (Kotlin/Java .aar), native iOS framework (Swift), and React Native are all available on request — just email us.
FaceNova blocks printed photo attacks, digital screen replays (phone/tablet/laptop), and 3D masks. The passive liveness model analyses texture, depth cues, and moiré patterns from a single frame — no blinking or head-turning needed from the user.
You pay once — $10 for 1 app, $49 for up to 10 apps — and get a license key tied to your package name(s). It never expires and includes all future SDK updates. No subscriptions, no renewal reminders.
Usually within a few hours, always within 24 hours. You fill in the form with your package name, we generate the key and email it to you along with a payment link if it's a paid plan.
No credit card. No commitment. We'll email your license key within 24 hours.
We'll email your license key and a payment link within 24 hours.
Pay $10 once, use forever — tied to 1 app package name. We'll email your key and payment link within 24 hours.
Pay $49 once, use forever across up to 10 package names. We'll email your key and payment link within 24 hours.
Last updated: January 2026
FaceNova SDK processes all biometric data entirely on the user's device. We do not collect, transmit, or store facial images or biometric templates on any server.
The FaceNova SDK operates entirely on-device. It does not send any face images, liveness scores, or biometric embeddings to FaceNova's servers or any third party.
result.faceTemplate are your responsibility to store securely within your own infrastructure.License keys are validated via Firebase Remote Config. This process sends your app's package name to Firebase to verify the license — no user biometric data is involved. See Google's Firebase Privacy Policy for details on how Firebase handles this data.
The SDK includes no analytics or crash reporting. Your app may independently use Firebase Crashlytics or similar tools — those are governed by their own respective privacy policies.
When you request a license through facenova.dev, we collect:
This data is used solely to issue and manage your license key. It is not sold or shared with third parties. We use Web3Forms to process form submissions; see their privacy policy for details.
FaceNova SDK is designed to support GDPR compliance. Since all biometric processing occurs on-device and no biometric data is transmitted to FaceNova, you retain full control over your users' data. As the app developer integrating the SDK, you are the data controller for any face templates you choose to store.
We retain your contact information (name, email, package name) for as long as your license is active plus 12 months. You may request deletion at any time by emailing privacy@facenova.uk.
FaceNova SDK is intended for use in apps targeting users aged 13 and over. If your application targets children, you are responsible for ensuring your use of the SDK complies with applicable children's privacy laws (COPPA, GDPR-K, etc.).
For privacy-related questions or data deletion requests, please contact: privacy@facenova.uk
Last updated: January 2026
Upon purchase or trial activation, FaceNova grants you a non-exclusive, non-transferable license to use the FaceNova SDK ("Software") within the single application identified by the package name specified at the time of purchase.
Trial licenses are valid for 10 calendar days from the date of activation and are provided for evaluation purposes only. Commercial or production use requires a paid license.
Monthly licenses are billed at $1 USD per month per app package name. Licenses renew automatically and may be cancelled at any time. Access continues until the end of the then-current paid period.
A Lifetime (Single) license grants perpetual access to the current version of the SDK and all future updates for 1 app package name, for a one-time fee of $10 USD. A Lifetime Bundle license covers up to 10 app package names for a one-time fee of $49 USD. "Perpetual" means the license does not expire as long as you comply with these Terms.
FaceNova will make commercially reasonable efforts to provide updates that maintain compatibility with current Flutter and OS versions. Lifetime license holders receive all future SDK updates at no additional cost. Monthly license holders receive updates while their subscription is active. FaceNova reserves the right to modify pricing for new license purchases.
The SDK, including all TFLite model files, encryption, and source code, remains the exclusive property of FaceNova. These Terms do not grant you any rights to trademarks, service marks, or other branding of FaceNova.
THE SDK IS PROVIDED "AS IS" AND "AS AVAILABLE" WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. FACENOVA DOES NOT WARRANT THAT THE SDK WILL MEET YOUR REQUIREMENTS, OPERATE WITHOUT INTERRUPTION OR ERROR, OR THAT ANY DEFECTS WILL BE CORRECTED.
TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, IN NO EVENT SHALL FACENOVA BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, OR PUNITIVE DAMAGES. FACENOVA'S TOTAL LIABILITY ARISING OUT OF OR RELATED TO THESE TERMS SHALL NOT EXCEED THE AMOUNT PAID BY YOU FOR THE LICENSE IN THE THREE MONTHS PRECEDING THE CLAIM.
FaceNova may terminate your license immediately upon written notice if you materially breach these Terms and fail to cure such breach within 14 days of notification. Upon termination, you must cease all use of the SDK and delete all copies in your possession.
These Terms shall be governed by and construed in accordance with the laws of England and Wales, without regard to its conflict of law provisions. Any disputes shall be subject to the exclusive jurisdiction of the courts of England and Wales.
For licensing or legal questions: support@facenova.uk