This application combines on-device open-vocabulary detection with an interactive frontend to auto-collect “snaps” (images + metadata) under configurable conditions.
It runs YOLOE on the DepthAI backend, and exposes controls in the UI for:
- Selecting labels (by text or image prompt)
- Adjusting confidence threshold
- Enabling snap conditions (timed, no detections, low confidence, lost-in-middle)
Note: RVC4 standalone mode only.
- Class control
- Update classes by text or upload an image to create a visual prompt
- Confidence filter
- Drop detections below a chosen threshold
- Snapping (auto-capture)
- Timed (periodic)
- No detections (when a frame has zero detections)
- Low confidence (if any detection falls below threshold)
- Lost-in-middle (object disappears inside central area; edge buffer configurable)
- Cooldowns reset when snapping is (re)started
Running this example requires a Luxonis RVC4 device connected to your computer. Refer to the documentation to set up your device if you haven't already.
To run the example in this mode, first install the oakctl tool using the installation instructions here.
The app can then be run with:
oakctl connect <DEVICE_IP>
oakctl app run .Once the app is built and running you can access the DepthAI Viewer locally by opening https://<OAK4_IP>:9000/ in your browser (the exact URL will be shown in the terminal output).
- You can upload oakapp to Luxonis Hub via oakctl
- And then you can just remotely open App UI via App detail