git clone git@github.com:xiaoBOSS97/handeye_calibration.git
cd handeye_calibration
pip install -r requirements.txt
-
store camera calibration images in
/eye2hand_cam/img_1
as png files, around 20-30 images -
store hand eye calibration images in
/eye2hand_cam/img_2
as png files, around 20-30 images (better different from camera calibration images) -
store the corresponding robot arm poses in
/eye2hand_cam/img_2/pose.npy
as npy files, example format [x, y, z, rx, ry, rz]
python main.py
- the images readed by
glob.glob()
are in random order, so you need to check the order of the images and poses in thepose.npy
file, and make sure the order is correct.