🏢

Onboarding

Provide the following details to add a new location and set up cameras.

📍

Location Details

Location Name *

Address *

0/250

City *

State *

Pincode *

Latitude (Optional)

Longitude (Optional)

🎥

Camera Details

Number of Cameras *

Camera Types *

e.g. Fixed, PTZ, ANPR, etc.

💡

Additional Use Case (Optional)

Select applicable use cases

👤

Point of Contact

Name *

Contact Number *

Email (Optional)

Designation (Optional)

Let's Make Your City Safer

Add your location and camera details to enable AI-powered monitoring and create a safer environment for everyone.

ℹ️

Need Help? You can modify these details later from Device Management.

Loading…

DeepStream Pipeline

Cameras, plate detector and TensorRT engine on the Jetson — edit, save and run from here.

Streams—configured

Average FPS—per stream

Output frame—auto-detected

Model · Engine——

Streams

RTSP cameras or video files. Resolution and fps are detected from each source.

#	Camera name	Type	Source URL	Detected	On

Detector model

Pick the YOLO .pt model. Its dynamic ONNX and the FP16 TensorRT engine for this number of cameras are made automatically and reused next time.

Model (.pt)

Engine size

—

Always the number of enabled cameras. An engine is reused only for exactly the same number.

Detection

How sure the detector must be, and how often it runs.

Plate confidence threshold

0.45 = 90% precision on our CCTV test set. Lower finds more small plates and more false boxes.

Detection interval