-
Thanks for your great work!!!!
I have a question about S3DIS. As we known, SEGColud just train on area 1-4,and 6, test on area5. However, others use six cross folder train/test. So the results are …
-
Hello @yunshin
Thank you for your research and allowing your code to be open sourced.
Currently, I tried to reproduce **SphericalMask** with the **S3DIS** dataset.
Since **SphericalMask** is …
-
Hi @HuguesTHOMAS ,
I am trying to run semantic segmentation using your code on a custom dataset which is similar to S3DIS but has fewer number of classes.
But the dimensions of the areas are muc…
-
Thank you for your excellent work!
I built a custom dataset modeled after the format of the s3dis dataset and got good results on the IoU and accuracy values for the semantic segmentation task, but h…
-
Is there any chance you could provide the actual S3DIS dataset after it has been segmented by various of the segmentation methods such as your PointNet++. PointNext, ect? I am hoping to compare how po…
-
hi, Bo, I have trained and eval on s3dis data, how can l visualization results? and I saw you have did it in (6) qualitative results on S3DIS
-
When will the S3DIS config for PointTransformerV3 be released?
-
![add5a778-9fc4-450d-ba92-02328516830e](https://github.com/Pointcept/Pointcept/assets/104781102/33357166-d890-4a24-8cdc-4a377165eea2)
In the OACNNs training code in the readme file, only the training…
-
亲爱的胡老师:
您好,我是一名本科生,最近在看您的文章,在运行的时候“sh jobs_6_fold_cv_s3dis.sh”这一条命令时,遇到了
“***EPOCH 0****
Caught a NaN error :
3
2 root error(s) found.
(0) Invalid argument: ValueError: attempt to get…
-
Thanks for your work. I found that there is a description "performance on 3Dfront->s3dis is quite unstable due to its simplicity and small sample sizes". I want to know if it can be improved by large …