RandLA-Net

RandLA-Net copied to clipboard

I want a pre_trained model of senmantic3D dataset,can you give me?

4

I want a pre_trained model of senmantic3D dataset,because my GPU is RTX3060,I can't us batchsize=4, I try to use batchsize=2,but the volatility is too great in train progress,not steable. my...

yangfeifan-ufo

> 你好，请问下现在Sementic3D官网能测试嘛，我账号审核通过了，但是提交测试的时候显示格式什么的都没问题，就是一直收不到结果邮件。 _Originally posted by @LiXinZhana in https://github.com/QingyongHu/RandLA-Net/issues/118#issuecomment-1196336393_

gsc1029

s3dis使用你提供的预测result可视化效果差

7

你好，我按照你提供的网盘链接https://drive.google.com/drive/folders/1tUov0iY88CWYAA0IP4nkOWCii0AN8Hem下载了预测结果，然后运行python utils/6_fold_cv.py，可视化发现预测的分割效果很差，地面存在各种颜色，墙壁也是。另外指标也很低。 下面该图是labels=original_data['class'] groud-truth的效果  下面该图是预测的效果。  下面该图是最终的评价指标结果  请问是哪里有问题导致效果跟你出入很大？

kkzh2313

pip install pandas==0.25.3 is not work `ERROR: Could not find a version that satisfies the requirement pandas==0.25.3 ERROR: No matching distribution found for pandas==0.25.3`

ZGX010

Toranto Problem

1

Hi, I noticed that you have detailed experiments and analysis on Toranto3d in the earlier submitted version. Can you disclose the corresponding preprocessing, training and testing code?

LujZhang


1981672112

OA and MIoU

1

In which log are the OA and MIoU of the training results in the paper seen? Epoch 0, end. Min possibility = 3.7 Confusion on sub clouds 56.33 | 90.16...

hhhhsmx

Can you give me RandLA-Net for trainning SensatUrban?

xsf777

Problem with qualitative result

1

I am trying to use your library in the SemanticKitti dataset. But I am troubled by the visualization problem. I want to compare the predicted result with the real label,...

Zhen-ao

Hi ; Please I would like to know how I can add other attributes (other information) in addition to the XYZ coordinates and RGB data, what are the parts to...

ayoubOUSSM2021