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
Thanks for your great work!
As we know, KITTI evaluate the result in BEV. But 'evaluate.py' and 'analyze.py' evaluate the result in perspective view. Could you tell me how can we evaluate in Bird's eye view locally(instead of uploading to KITTI server)?
The text was updated successfully, but these errors were encountered:
Similar question. KITTI provides the TF-tree and calibration file for perspective transformation. So I wish I could extract the pixel coordinates of the road part in image with modifying KittiSeg codes and transform them into the coordinates in the base_link (car coordinate system). Please let me know if you have any ideas. Thanks.
Thanks for your great work!
As we know, KITTI evaluate the result in BEV. But 'evaluate.py' and 'analyze.py' evaluate the result in perspective view. Could you tell me how can we evaluate in Bird's eye view locally(instead of uploading to KITTI server)?
The text was updated successfully, but these errors were encountered: