You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Given a rgbd videos, I want to run ScrewNet and generate the predicted rotation axis. I've checked the code and it only gives me the number. How do I generate predictions like Fig 7 (real world image) in the paper? Thank you very much!
The text was updated successfully, but these errors were encountered:
ScrewNet predicts the screw axis in the object's local frame of reference, which is placed at the center of the bounding box of the object. If you want the axis in the camera frame, you will additionally need the object's pose in the camera frame. With that, you can use equation 2 from the paper to define a transformation matrix for the screw axis and transform the axis to the camera frame (You can find an implementation of it in function "utils.py/transform_plucker_line" in the repo at: https://github.com/Pearl-UTexas/ScrewNet/blob/main/utils.py#L225 ). You can then overlay the screw axis in the camera frame on the image using standard opencv functions (cv2.projectPoints). Please let me know if you have any questions.
Given a rgbd videos, I want to run ScrewNet and generate the predicted rotation axis. I've checked the code and it only gives me the number. How do I generate predictions like Fig 7 (real world image) in the paper? Thank you very much!
The text was updated successfully, but these errors were encountered: