How to use it
- Download the pretrained YOLOv3 weights.
- Record or provide a video.
- Run the script and see detected objects.
How it's built
YOLOv3 with OpenCV reads video frame by frame and detects object names, similar to a managed cloud recognition service.
Why it matters for businesses
Local video analysis avoids per-minute cloud costs and keeps footage private.
In real life
A shop counts how many people and bags pass the entrance each hour, without uploading video anywhere.


