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

Pre-trained models for monocular depth networks #15

Closed
VitorGuizilini-TRI opened this issue Jun 1, 2021 · 5 comments
Closed

Pre-trained models for monocular depth networks #15

VitorGuizilini-TRI opened this issue Jun 1, 2021 · 5 comments

Comments

@VitorGuizilini-TRI
Copy link

Thank you for open-sourcing this very interesting work. Would it be possible to also provide weights for the monocular depth networks (the teacher networks) that go along with the currently available pre-trained models? Thank you!

@JamieWatson683
Copy link
Collaborator

Hi - sorry for the delay.

Sure, I will try to dig them out and upload.

@JamieWatson683
Copy link
Collaborator

Hi - just added to the zips for the KITTI models.

Let me know if you need the cityscapes one too and I can add.

Thanks a lot!

@JamieWatson683
Copy link
Collaborator

Additionally I've added some code to evaluate the teacher networks themselves - I hope this helps!

@VitorGuizilini-TRI
Copy link
Author

Thank you! Do you know if these teacher networks, when trained from scratch jointly with the student multi-frame network, achieve a similar performance to when they are trained separately, as in monodepth2?

@JamieWatson683
Copy link
Collaborator

Yes, they generally perform to the same level as monodepth2 - I've added the ability to test these by adding thee flag --eval_teacher to the evaluate_depth.py script. Let me know if that helps!

Thanks a lot

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

2 participants