search

TRAILab / CaDDN

Categorical Depth Distribution Network for Monocular 3D Object Detection (CVPR 2021 Oral)
Apache License 2.0
359 stars 62 forks source link

Questions about MaptoBEV operation #42

Closed gujiaqivadin closed 3 years ago

gujiaqivadin commented 3 years ago

Hello! Thanks for your sharing great mono3d detection work CaDDN! After I read the code, I have a simple question about the Map-to-BEV operation. I found two operations here 'Conv2DCollapse' and 'HeightCompression', and they do the same thing to collapse voxel-wise features into BEV features. I want to know where is the difference between these two operations. And the code of reshape operation '.view' and '.flatten' , is there any difference between these two ways?

Thanks a lot!

codyreading commented 3 years ago

Hi and thanks for the interest!

The only real difference between these is that Conv2DCollapse includes a 1x1 convolutional block for channel reduction. Otherwise, they should function the same. Reshaping via .view or .flatten ends up being the same.

gujiaqivadin commented 3 years ago

Thanks a lot!