search

SysCV / LiDAR_snow_sim

LiDAR snowfall simulation
https://www.trace.ethz.ch/lidar_snow_sim
Other
177 stars 28 forks source link

What's the use of precompute.py #12

Closed Rolandxx7 closed 2 years ago

Rolandxx7 commented 2 years ago

Thanks for your great work! I wanna know what is the purpose of this file called precompute.py, because I do not see any corresponding note.

MartinHahner commented 2 years ago

With precompute.py, you can compute the snowflake patterns, which are then stored on disk. This process in its current implementation takes too much time to do on the fly while training your model. During training, you can then load those snowflake patterns from the disk without computing them each time.

Rolandxx7 commented 2 years ago

With precompute.py, you can compute the snowflake patterns, which are then stored on disk. This process in its current implementation takes too much time to do on the fly while training your model. During training, you can then load those snowflake patterns from the disk without computing them each time.

Thanks for your kind reply! Also I would like to know why do you use calibration of kitti dataset instead of dense dataset in precompute.py. Because you have said you

With precompute.py, you can compute the snowflake patterns, which are then stored on disk. This process in its current implementation takes too much time to do on the fly while training your model. During training, you can then load those snowflake patterns from the disk without computing them each time.

Thanks for your kind reply! Also, I would like to know why you use the calibration of kitti dataset instead of dense dataset in precompute.py. Because you have said that the code only implemented the snow simulation of dense dataset

MartinHahner commented 2 years ago

We use the calibration file from the DENSE dataset. It is compatible witht the KITTI calibration code. So we just re-use the KITTI code, but not their calibration.