LPCG

LPCG copied to clipboard

process data meet error

4

hello， i run ‘ python data/kitti/link_kitti_raw.py’，meet error， Traceback (most recent call last): File "data/kitti/link_kitti_raw.py", line 40, in os.symlink(img_list[i], os.path.join(image_dst_dir, base_name+'.png')) FileExistsError: [Errno 17] File exists: './data/kitti/raw_data/2011_09_28/2011_09_28_drive_0001_sync/image_02/data/0000000035.png' -> './data/kitti/kitti_merge/training/image_2/000000.png' （I did...

Watebear

Some questiones about low_cost mode

2

@SPengLiang 在low_cost 模式下，我理解我是在完全没有标签的情况下对baseline进行训练对吧。请问是否可以将low_cost扩展为这样的模式：对于kitti train数据集中3700+有标签数据，我使用人工标签，对于kitti raw中未标记的数据我使用low_cost生成的标签？现在我似乎是这样的，无论一幅图像是否有对应的人工标记，我在low_cost处理后都会使用伪标签？ 请问是不是high模式下，才会使用到人工标签，对于半监督的逻辑我有一些疑惑，谢谢您？

aqr1961

Support for Waymo

1

Hi there, can you please provide necessary codes to support experiments on Waymo?

carlvinson70

TypeError: list indices mustbe integersor.slices,nottuple

2

File "/media/10T/weakm3d/dataloader/kitti dataloader.py", line' 152, in __getitem__ cls info = np.ones like(RoI boxpoints['bbox2d'][:,0])*2 TypeError: list indices mustbe integersor.slices,nottuple 请问是否有人遇到过类似的问题，怀疑是遇到了点云为空的情况，是否可以尝试利用判断语句直接去掉空点云呢？

aqr1961

how to use the new label by created low cost

1

Hi, Thanks for your great work, I have a question about labels for the train set : 1、I have used low mode to generate labels, but I cannot find the...

lzh111222334

How to generate validation labels for low cost mode?

1

Hello, Thanks for your great work，I have a question about the validation labels for low cost mode： 1.How to generate validation labels and validation image for low cost mode? I...

870181985

Labels for train set

2

Hi, Thanks for your great work, I have a question about labels for the train set : 1. All all the samples in the train set included in the raw...

Yinyf0804

[Errno 2] No such file or directory: './data/kitti/kitti_merge/training/calib_cam2cam/000000.txt'

3

I meet some troubles, can you give me some advice? thanks ------------------------------------------------------------------------ 100%|██████████| 33530/33530 [00:04

870181985

Hi! is there possible to apply on custom dataset? I have custom dataset with the KITTI format and want to try the low cost mode

AlifyaFebriana