Face AI

Tatvora Face Recognition AI

Recognise identities using intelligent facial analysis.

Tatvora Face Recognition AI performs one-to-many identification. Your application sends a captured face together with the candidate identity set to compare it against, and the API returns the matching identity where one is found. It is the right model when your application needs to answer "who is this?" rather than confirm a claimed identity. The call is stateless: your enrolled identities stay in your database, we hold no face gallery, and nothing from the request is retained after the response.

POST https://api.tatvora.com/api/v1/face/recognize
POST https://api.tatvora.com/api/v1/face/recognize
curl -X POST https://api.tatvora.comPOST /api/v1/face/recognize \
  -H "Authorization: Bearer $TATVORA_API_KEY" \
  -F "image=@capture.jpg" \
  -F "candidates=@candidates.json"
What this model does — and does not — do Face Recognition performs one-to-many identification against the candidate set you send with the request — we do not host an enrolled gallery, so your identities never leave your systems. To confirm that two specific images are the same person, use Face Verification instead.
How it works

The Face Recognition pipeline

  1. Captured face + your identity set
  2. Tatvora Face Recognition AI
  3. Embedding comparison
  4. Matched identity
See it work

What you send, and what you get back

A worked example of a single Face Recognition AI call, end to end.

POST /api/v1/face/recognize

1 · You send

+ your candidate set

2 · We process

Tatvora Face Recognition AI 1:N comparison

3 · You receive

Employee B recognised
Matched Yes
Identity emp_10428
Similarity 0.91
Candidates compared 3
{
  "success": true,
  "matched": true,
  "identity_id": "emp_10428",
  "similarity": 0.91,
  "candidates_considered": 3
}

Sample data for illustration. Your enrolled identities are sent with the request and stay in your own database — Tatvora holds no face gallery, so there is no biometric store on our side. Set the similarity threshold your use case requires.

Capabilities

What Face Recognition AI can do

  • Identify a person from a captured face against a supplied candidate set
  • Accept candidates as images or as embeddings returned by a previous call
  • Facial feature analysis and embedding comparison
  • One-to-many recognition across the identity set you provide
  • Return the matched identity with a similarity score
  • Stateless by design — no face gallery is held on our side
Use cases

Where it is applied

  • Employee attendance and time capture
  • Physical and logical access control
  • Visitor management
  • Workforce management platforms
  • Member and customer identification
  • Recognising returning users in an application
  • Secure internal applications
API reference

Request and response

Authenticate with a bearer API key over HTTPS. Upload two files as multipart/form-data and receive a JSON response.

Request fields

  • image — file part, multipart/form-data
  • candidates — file part, multipart/form-data
  • Authorization: Bearer YOUR_API_KEY — required header

Endpoints and payloads shown on this site are illustrative examples of the published interface. Never expose a secret API key in client-side code.

200 OK — example response
{
  "success": true,
  "request_id": "req_5c04ee18",
  "matched": true,
  "identity_id": "emp_10428",
  "similarity": 0.91,
  "candidates_considered": 1240
}
FAQ

Face Recognition AI questions

Your application sends the captured face and the candidate identity set together in the same request, either as images or as embeddings returned by an earlier call. We compare them, return the match, and discard both. Your enrolled identities stay in your own database, which means there is no Tatvora-side biometric gallery to breach or to ask us to erase.

Face Detection answers "is there a face here, and where?" and returns bounding boxes. Face Recognition answers "who is this?" by comparing a captured face against the candidate identity set your application sends with the request. Detection carries no identity information.
More AI APIs

Combine Face Recognition AI with the rest of the platform

Document AI

Tatvora OCR AI

Turn documents into usable data.

Extract text and structured information from images, scans, PDFs and office documents through a single document intelligence API.

View all AI products

Try Tatvora Face Recognition AI

Request an evaluation key and validate the model against your own data before choosing a plan.

Questions? Email support@tatvora.com