Understanding 1:N Face Recognition and 1:1 Face Verification
Learn the difference between 1:N face search and 1:1 face verification — how each works at the API level and when to use which approach.
1:N vs 1:1 — What Do the Numbers Mean?
In biometrics, "1:N" and "1:1" describe how many comparisons are made:
For a detailed breakdown of all face technologies, see our guide on face detection vs. recognition vs. verification.
1:N Face Recognition
You register faces into a database, then send a new photo to find matches:
Register a face
requests.post(
"https://faceapi.arsa.technology/api/v1/face_recognition/register_face",
headers={"x-key-secret": KEY, "x-face-uid": "employee_42"},
files={"face_image": open("photo.jpg", "rb")}
)
Later — recognize who this is
response = requests.post(
"https://faceapi.arsa.technology/api/v1/face_recognition/recognize_face",
headers={"x-key-secret": KEY},
files={"face_image": open("unknown.jpg", "rb")}
)
Returns: {"recognition_uidresult": "employee_42", "recognition_confidence": 0.95}
Best for: Attendance systems, access control, visitor management.
1:1 Face Verification
No database needed — compare two specific images directly:
curl -X POST "https://faceapi.arsa.technology/api/v1/face_recognition/validate_faces" \-H "x-key-secret: YOUR_API_KEY" \
-F "image1=@selfie.jpg" \
-F "image2=@passport_photo.jpg"
Best for: KYC onboarding, transaction authorization, hotel check-in.
Choosing the Right Approach
Use 1:N when you need to identify someone from a group. Use 1:1 when you need to confirm someone is who they say they are.
ARSA supports both in the same API. Get started free or explore the docs.