Deploy & Test

Upload your trained YOLO model, connect a webcam or video, and see real-time detections with persistent tracking.

Connect GPU

Using the default Falcon Perception backend configured in your DLF server.

25%

Start webcam or upload a video to begin detection

Frames are sent for inference, bounding boxes drawn in real time

Idle

FPS

0.0

Latency

0ms

avg 0ms

Detections

0

avg 0

Frames

0

Total Tracks

0

Active Now

0

Model

No model uploaded

Using default backend inference

Backend

EndpointFalcon (local)
TypeDLF Default
Queryobject
Threshold25%

How It Works

  1. Upload a trained YOLO model (.pt file)
  2. Start your webcam or upload a video file
  3. Each frame is sent to the inference backend
  4. Bounding boxes are drawn with persistent IoU tracking
  5. Connect a RunPod GPU endpoint for faster inference