Capabilities
Faceplugin Face SDK capabilities: face recognition (1:1, mobile 1:N Identify), standalone face liveness (PAD), and combined Recognition + Liveness on port 8083.
What the Faceplugin Face SDK can do — recognition, liveness, and recognition + liveness — before you pick a platform.
Recognition
After you activate with a license key, face detection, template extraction, and matching run offline on the device or on your Windows/Linux host. Images are not sent to a Faceplugin cloud.
The algorithm is evaluated on NIST FRVT (as stated across Faceplugin Face Recognition docs).
1:1
Mobile and server
Extract two templates → similarity / POST /api/similarity, or POST /api/match with two images
1:N Identify
Mobile demos
VideoWorker + templates you store in your database (demo default match threshold 0.67)
On Linux / Windows, the HTTP API is still-image detect / quality / feature / match / similarity. There is no server-side POST /api/identify gallery — you store templates and call similarity yourself.
templateExtraction / POST /api/feature generates face templates that you can store in your own database. Faceplugin does not host your users' face data or provide a cloud-based person gallery.
Mobile 1:N Identify includes passive 2D liveness (default threshold 0.5) as part of the identify flow. That is not the same as the standalone Liveness product below.
Recognition API (port 8083)
GET /api/machinecode
Machine code
POST /api/activate
Activate
POST /api/detect
Detect faces
POST /api/quality
Quality
POST /api/feature
Template
POST /api/match
Two images
POST /api/similarity
Two templates
Image: faceplugin/face-recognition. See Linux · Windows.
Liveness (standalone PAD)
Standalone passive presentation-attack detection: scores a camera frame or JPEG without a smile / turn-head challenge. These docs do not claim iBeta certification. Contact Faceplugin if you need a certification statement.
Detects attack types including:
Printed photos
Screen replays
3D models / masks
Deepfake-style video (one type of spoof — not a separate deepfake product)
Score ≥ 0.5 → Real / pass (server default threshold in these docs).
Android / iOS
Live camera + VideoWorker / faceDetection for anti-spoofing
Linux / Windows
POST /api/liveness with one RGB JPEG on port 8084
There is no public Face Liveness Flutter or React Native SDK. Use native Liveness, or 2D liveness on Face Recognition Identify.
Images: faceplugin/face-liveness. See Linux · Windows.
Recognition + Liveness (one server process)
One container / process on port 8083: all recognition routes plus POST /api/liveness. Image: faceplugin/face-recognition-liveness-sdk.
Which mode to pick
Enroll / 1:N match (mobile Identify includes 2D liveness)
Recognition
Anti-spoofing only (no gallery)
Liveness
Recognition + anti-spoofing in one server process
Recognition + Liveness
Document authenticity (ID spoof)
Platforms
Use cases
Access control / attendance — enroll templates; live 1:N Identify
KYC selfies — passive anti-spoofing before or with face match
Server fraud checks — still-image detect / match / similarity on 8083, or liveness on 8084 / combined 8083
FAQ
Can face recognition work completely offline? Yes, after license activation.
Is there a Face Recognition API? Yes — HTTP on port 8083.
Server-side 1:N gallery? No. Store templates in your DB; call /api/similarity.
Difference between Identify 2D liveness and Face Liveness? Identify embeds 2D liveness in matching; standalone Liveness is anti-spoofing only (port 8084).
Related documentation
Last updated