search

xiaoaoran / SemanticSTF

(CVPR 2023) The official project of "3D Semantic Segmentation in the Wild: Learning Generalized Models for Adverse-Condition Point Clouds"
95 stars 12 forks source link

Question about SemanticKITTI→SemanticSTF Evaluation #1

Closed at7am closed 1 year ago

at7am commented 1 year ago

Hi, and thanks for sharing the good work!

I had a general question regarding the evaluation setup that you show in Table 2 of the paper. The question is, when you train your model on the SemanticKITTI, you classify points in one of the 19 classes (car, ..., traf.). However, the SemanticSTF has also the 4 adverse weather classes in addition to the 19 base classes.
How do you segment them if your model outputs only 19 logits?

The same goes of the SynLiDAR→SemanticSTF evaluation.

xiaoaoran commented 1 year ago

Hi, @at7am

We separate data according to weather type and evaluate miou. Please refer to 'evaluate_by_weather.py'

at7am commented 1 year ago

Hi, thanks for your answer. So basically the table shows the class-wise IoU for the entire dataset, and then the mIoU when segmenting in a particular weather condition?

xiaoaoran commented 1 year ago

Yes.