- Clone the following repository:
git clone https://github.com/iamlab-cmu/perception.git
cd perception
- Run the following in a virtual environment:
- Clone this repository:
git clone [email protected]:iamlab-cmu/camera-calibration.git
cd camera-calibration
- Run the following in a virtual environment:
- Start the Azure Kinect Camera
roslaunch azure_kinect_ros_driver driver.launch
- Open the calib/azure_kinect.intr file.
- Rostopic echo /rgb/camera_info once.
- Copy the first number in K: to after "_fx"
- Copy the third number in K: to after "_cx"
- Copy the fifth number in K: to after "_fy"
- Copy the sixth number in K: to after "_cy"
- If you are using the large checkerboard for an overhead camera, just run:
python scripts/register_camera.py
- If you are using a checkerboard on the hand, first move the robot into a position so that the checkerboard can be seen from the camera at a distance of around 2 feet away.
- Then start franka-interface and use the following command:
python scripts/register_camera_using_franka.py
- Afterwards, check to make sure the calib/azure_kinect_overhead/azure_kinect_overhead_to_world.tf file is similar to calib/example_azure_kinect_overhead_to_world.tf.
- If the transformations are similar, then test the calibration using a flat surface and an alphabet block and run the following command:
python scripts/run_pick_up_using_camera.py
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent