Powerful biometrics as a service.
Enrich your apps.
Fast and easy integration.
Solutions for face.com developers
The popular face recognition API from face.com was recently shut down after their acquistion by Facebook.
BioID Web Service has a face recognition API that could be a viable alternative for some former face.com developers.
It's not identical; there are features face.com had that BioID doesn't support, but BioID can also do things face.com couldn't.
- BioID technology was developed for authentication applications: recognition is far more accurate than typical for photo tagging.
- BioID provides liveness detection, to detect impostors trying to authenticate as someone else by holding a photo or video in front of the camera.
- Face detection is more strict (necessary for authentication). This means it will not find faces that are heavily turned or tilted, or extremely small/extremely low resolution.
- BioID can return the eye position coordinates and/or a cropped image of just the face, but not any further coordinates for face position, nor gender, glasses, etc.
- Multiple faces: for authentication BioID detects and rejects images with multiple faces, to avoid looking over the shoulders scenarios. For other applications, with multiple faces BioID can return the eye coordinates and/or a cropped image for the best face found.
- BioID is currently available as a SOAP API, a RESTful version is planned.
- No usage limits.
- 30-day free trial, then pricing on a per-transaction basis based on an annual subscription.
- Biometric authentication (face & voice; iris coming soon), including liveness detection
- Biometric identification (one-to-many): identifying people on blacklists, criminal background checks
- Ticketless check-in: trade shows and other events, airline boarding, ships
- Fraud prevention: liveness detection to verify presence of a live user; detecting multiple registrations from the same person
- Face or voice similarity search: celebrity matching, online dating, modeling or casting agencies
- Photo quality checks to ensure good-quality (biometric) ID photos or mugshots
- Limited photo tagging when there is only one face in the image and the face is turned relatively straight towards the camera
- Recognizing multiple faces in one image
- Finding or recognizing heavily turned or tilted faces
- Identifying attributes such as gender, age, wearing glasses, smiling, etc.
- Returning location of features other than eyes