Closed chowkamlee81 closed 4 years ago
Sorry for the delay.
We used https://github.com/traveller59/second.pytorch to train on the KITTI data. The model is then applied to the KITTI scans to have the bounding box. The semantic segmentation is generated by your favorite model.
For merging the predictions, we use the following script: https://gist.github.com/jbehley/09591d1d8db400a5afc4a63eaeb026cd
Thanks got the idea. Closing the issue
Would like to know baseline code for panoptic segmentation on semantic kitti