Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

transform parameters #1

Open
RG2806 opened this issue Aug 24, 2022 · 4 comments
Open

transform parameters #1

RG2806 opened this issue Aug 24, 2022 · 4 comments

Comments

@RG2806
Copy link

RG2806 commented Aug 24, 2022

Hey, there
great work!
I wanted to know what are these parameters and how are they set?

oheight, owidth, cwidth = 256, 1216, 1216

@LittlePey
Copy link
Owner

The sizes of KITTI images are different, to be convenient, we crop the bottom-center (256 x 1216) of each image and feed it to the depth completion network.

def test_transform_(rgb, sparse, target, args):

@RG2806
Copy link
Author

RG2806 commented Aug 25, 2022

i wanted to check results of your pre-trained model on my custom dataset, all images are size 40963000 (wh). What values should i use than?

@LittlePey
Copy link
Owner

Hi, you need to train TWISE on your custom dataset in advance, it is not recommended to use TWISE pre-trained on KITTI depth completion dataset to inference on your custom dataset due to the domain gap.

@longsuo
Copy link

longsuo commented Jun 26, 2023

Hi, you need to train TWISE on your custom dataset in advance, it is not recommended to use TWISE pre-trained on KITTI depth completion dataset to inference on your custom dataset due to the domain gap.

But the training code has not been provided.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants