search

maudzung / SFA3D

Super Fast and Accurate 3D Object Detection based on 3D LiDAR Point Clouds (The PyTorch implementation)
https://github.com/maudzung/Super-Fast-Accurate-3D-Object-Detection
MIT License
1k stars 271 forks source link

2D pcd file testing #71

Open ncusspm25 opened 2 years ago

ncusspm25 commented 2 years ago

Thank you for such a good work. I want to train my own 2D lidar pcd without calib and image file. I have transfered the format of my label file and bin file to kitti dataset. But when I train successfully, test.py can't detect my item. I think it's related to bev or some other issue. Which file should I modify? Thank you very much!

yukun093 commented 2 years ago

Thank you for such a good work. I want to train my own 2D lidar pcd without calib and image file. I have transfered the format of my label file and bin file to kitti dataset. But when I train successfully, test.py can't detect my item. I think it's related to bev or some other issue. Which file should I modify? Thank you very much!

Hi, have you solved your problem to train your own dataset?

mhmodayman commented 1 year ago

Thank you for such a good work. I want to train my own 2D lidar pcd without calib and image file. I have transfered the format of my label file and bin file to kitti dataset. But when I train successfully, test.py can't detect my item. I think it's related to bev or some other issue. Which file should I modify? Thank you very much!

How did you train on the data without calib and image?

martinteh commented 1 year ago

Did you manage to parse/test with your own data? image

My point cloud exists, but for some reason is not showing Can't seem to calibrate or get this working