occupancy-for-nuscenes

occupancy-for-nuscenes copied to clipboard

huge long time to generate trainval GT

4

Thanks for your work. But I find that it takes huge long time to generate trainval GT even with parallel generating. Also, it takes almost all CPU threads. Is that...

qihao-plus

Related Paper & Performance

2

Thanks for sharing this amazing work! I wonder, if there is a paper related to this project. Can you add a link if exists? Could you please provide some information...

YoushaaMurhij

请问如何定义不同的occupancy尺寸？

1

非常感谢你的代码，请问如何定义不同的occupancy尺寸？在data_converter中没有找到如何确定生成gt尺寸相关的设置，但是看你的train代码中定义的BEV尺寸是100x100

shabiouyang

# Welcome update to OpenMMLab 2.0 I am Vansin, the technical operator of OpenMMLab. In September of last year, we announced the release of OpenMMLab 2.0 at the World Artificial...

vansin

Pretty good work, can you provide your training log, I want to compare it with my log.

revisitq

" velocities = np.concatenate((velocities, np.zeros_like(velocities[:, 0:1])), axis=-1) IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed"

6

Hi, When I run the `data_convert.py` to generate the occupancy data, I meet the error below. ```` 3836 3837 Traceback (most recent call last): File "data_converter.py", line 464, in convert2occupy(args.dataroot,...

rockywind

How to deal with velocities=[]

1

Hi, when I run your code on the trainval set, in some scenes it will meet the following problem: velocities = np.concatenate((velocities, np.zeros_like(velocities[:, 0:1])), axis=-1) IndexError: too many indices for...

Robertwyq