Legacy example · Python · Version 1.0
USB Camera Calibration
| Authors | Holoscan Team (NVIDIA) |
|---|---|
| Supported platforms | aarch64 |
| Language | Python |
| Version | 1.0 |
| Compatibility | HSDK 2.2 |
| Contribution level | Level 1 · Very high-quality code |
| Source | HoloHub |
Legacy HoloHub application
This application comes from HoloHub and was last tested with Holoscan SDK (HSDK) 2.2. It runs with the HoloHub CLI, ./holohub.
Overview
This application performs monitor registration using a USB camera. It detects April tags placed at the four corners of a monitor to establish the monitor's position and orientation in 3D space.
Hardware Requirements
- Logitech 4k Pro Webcam or compatible USB camera
- Monitor with April tags at all four corners
- Proper lighting conditions (well-lit environment without backlight)
Setup Instructions
- Ensure the USB camera is properly connected
- Place the calibration image with April tags on the monitor
- Position the camera so it can see all four corners of the monitor
- Verify camera visibility using the v4l2_camera app
Running the Application
LD_PRELOAD=/usr/lib/aarch64-linux-gnu/nvidia/libnvjpeg.so \
./holohub run usb_cam_calibration --local
If device permissions require root, add --as-root (never sudo ./holohub).
Output
The application generates a calibration file usb-cali.npy in the build directory, which contains the monitor's corner coordinates.
Notes
- The camera must have a clear view of all four April tags
- Avoid backlighting or glare on the monitor
- If using a different camera model, update the camera settings in the Python app or YAML configuration file