dapengfeng / waymo-toolkit Goto Github PK

Hi, thank you for sharing your toolkit.

I tried to use your toolkit but it was slower than I expected.
It took 64 secs per frame to extract laser data with single TITAN RTX GPU.
Is it the same speed with yours?

By the way, thank you again for your work!

Running extractor.py with --image was successful and I extracted the images properly. However, I faced 2 issues with --label. Firstly, the number of labels extracted was a lot less than number of images. Secondly, all the labels extracted (in .pb files) were just descriptions of the cameras, without any information about the objects in each image. Could I clarify the reason for this?

Hi, How can i convert outputed label.pt to 2d boudingbox prediction txt?

Thanks

Traceback (most recent call last):
File "tools/extract.py", line 7, in
from waymo_toolkit.extractor import ImageExtractor, LabelExtractor, LaserExtractor
File "/home/hy/anaconda3/envs/waymo-toolkit/lib/python3.7/site-packages/waymo_toolkit-0.1.0-py3.7.egg/waymo_toolkit/extractor/init.py", line 1, in
File "/home/hy/anaconda3/envs/waymo-toolkit/lib/python3.7/site-packages/waymo_toolkit-0.1.0-py3.7.egg/waymo_toolkit/extractor/image.py", line 7, in
ModuleNotFoundError: No module named 'waymo_extractor'
I met this error, how to fix it?