Hello,
I have two questions regarding your work and would greatly appreciate your clarification:
In Table 2, Comparison of panoptic segmentation performance on the nuScenes validation set, is the mIoU metric computed in the same way as the mIoU used in conventional point cloud semantic segmentation? If not, could you please explain how the mIoU reported in the table is calculated?
Would it be possible to release the training configuration files used for the SemanticKITTI dataset, as well as the required dataset format and preprocessing specifications?
Thank you very much for your time and assistance. I look forward to your reply.
Best regards,
Hello,
I have two questions regarding your work and would greatly appreciate your clarification:
In Table 2, Comparison of panoptic segmentation performance on the nuScenes validation set, is the mIoU metric computed in the same way as the mIoU used in conventional point cloud semantic segmentation? If not, could you please explain how the mIoU reported in the table is calculated?
Would it be possible to release the training configuration files used for the SemanticKITTI dataset, as well as the required dataset format and preprocessing specifications?
Thank you very much for your time and assistance. I look forward to your reply.
Best regards,