Jens Behley

icon indicating a website link https://jbehley.github.io/

icon company University of Bonn icon location Bonn, Germany icon location Post-doctoral Researcher at the Department of Photogrammetry, University of Bonn.

Results 108 comments of Jens Behley

Problem using visualize.py

thanks for your interest in our work: I suspect that it has nothing to do with the GPU as it's only visualizing the point clouds; but I haven't checked in...

Problem using visualize.py

okay with an old version, it works?

Problem using visualize.py

Also here, it seems that the issue is either resolved or isn't relevant anymore.

Corresponding semantic masks on images

hi hugo, as far as I know there are now image labels that directly correspond to the LiDAR labels. However, I would guess that by projecting the point cloud labels...

Corresponding semantic masks on images

I think the issues seems resolved or at least answered. If you have still questions then don't hesitate to reopen the issue.

test label

Please note that we do not release the test labels, and you have to submit your predictions to the corresponding Codalab Challenges: https://semantic-kitti.org/tasks.html For each of the different tasks, we...

test label

I guess also this issues has been clarified. If you have further problems, then open a new issue or re-open this issue.

issue about data

Please see http://semantic-kitti.org/dataset.html#download, where we link to what to download and where. If you have you have further question, let me know.

issue about data

I guess the issue could be resolved, therefore, I close the issue now. If you still have problems, then just re-open the issue.

representation question of SemanticKITTI

Not sure what you mean; I would instinctively simply use a [0, ..., K] mapping for the K classes + unlabeled classes to keep the point cloud information. But probably,...