SUSTechPOINTS
SUSTechPOINTS copied to clipboard
3D Point Cloud Annotation Platform for Autonomous Driving
作者你好, 标注完成的KITTI格式数据,在OpenPCDet下训练报错. 而原始的kitti数据集没有问题. 报错信息如下: OpenPCDet/lib/python3.8/site-packages/torch/_utils.py", line 429, in reraise raise self.exc_type(msg) ValueError: Caught ValueError in DataLoader worker process 0. ... /OpenPCDet/lib/python3.8/site-packages/numpy/core/shape_base.py", line 460, in stack raise ValueError('need at least one...
Hi Naurril. May I know did you encounter this situation where the loading of pcd file and images just get super slow after a while (random, it could be one...
批量标注如何使用,readme中写的无法操作
有些pcd文件使用reflectance字段而不是intensity字段,本次pr增加了当intensity字段不存在时使用reflectance字段代替的功能
您好,我在calib/camera里面已经改好了两传感器的外参和相机的内参,但是点云并没有投影到图像上,我看到框的大小过大(超过了图像的范围),以及位置也有很大偏差,但是这个标定参数我之前在另一个项目使用过,是可以较为准确的将点云投影到图像的(autoware标定数据),我是需要再修改其他的设置文件吗?
如何将标注好的文件转为kitti格式的文件,/tools/sustechscapes_to_kitti.py 这里只有pcd2bin,标注信息如何转化
你好,我将一个包含200帧点云的文件夹进行了标注,前80帧都正常工作,但是当我读取第81帧时点云不在更新,我返回查看前80帧的数据时点云也不在更新,整个程序卡住了 我使用的是Chrome 122.0.6261.129版本,当我通过F12查看控制台的时候发现了如下问题: 
- [x] 建立Action: 每次发版时自动构建Docker镜像 (auto build docker image when release an updated version) - [x] 建立Action: 每次发起合并请求时, 自动触发代码风格检查, 语法检查( Trigger code format&syntax check when put on a pull request) -...
Could you provide me a python file for reading pcd data correctly within the /example/lidar folder? I have tried to program by myself and search related info, but all failed....
Hi @naurril , I hope you're doing well, I just want to ask you the format of Sustech 1 - is the "position" the center coordinate of the bbox? 2...