Expected mIoU of the pretrained checkpoint

xinge008 / Cylinder3D

Rank 1st in the leaderboard of SemanticKITTI semantic segmentation (both single-scan and multi-scan) (Nov. 2020) (CVPR2021 Oral)

Apache License 2.0

859 stars 180 forks source link

Expected mIoU of the pretrained checkpoint #44

Closed JBKnights closed 3 years ago

JBKnights commented 3 years ago

Hello,

Would you be able to tell us what the expected mIoU of the pretrained checkpoint you supply should be? In the environment I'm using I got a mIoU of 66.914, which is significantly higher than what you report in your paper. Some clarification on why that might be the case would be very helpful.

Thanks!

xinge008 commented 3 years ago

The model provided in the codebase performs slight higher than original paper with more training epochs.

owoshch commented 3 years ago

Hi @JBKnights! I'm trying to reproduce the results you posted. Could you please share the command you ran to obtain those results? Thank you!

GaloisWang commented 2 years ago

excuse me, when I test the pretrained model on KITTI 08 sequence, the results are all 0. Have you encountered this problem before？