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.

Sign up for GitHub

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

Jump to bottom

How can I use ViT? #525

Closed
Shiming94 opened this issue Apr 28, 2021 · 5 comments
Closed

How can I use ViT? #525

Shiming94 opened this issue Apr 28, 2021 · 5 comments
Assignees
@Junjun2016

Comments

@Shiming94
Copy link

Hi guys,

I have noticed that you have updated the VisionTransformer backbone, is there a network, that already uses it? I really want to see the performance of Transformer in Semantic Segmentation.
@clownrat6
Copy link
Contributor

The relative configs are benchmarking. We may support vit+[fcn/deeplab/pspnet/upernet]. Please wait a minute. Furthermore, we will support SETR soon~.

@jianlong-yuan
Copy link

any news

@clownrat6
Copy link
Contributor

If you are eager to use ViT for semantic segmentation, you can firstly refer to #520.

@Junjun2016
Copy link
Collaborator

Hi @jianlong-yuan and @Shiming94
We have released the SETR benchmark which is based on ViT.
And we will release other Transformer based semantic segmentation benchmarks ASAP.

@Junjun2016 Junjun2016 self-assigned this Jun 24, 2021
@Shiming94
Copy link
Author

Hi @jianlong-yuan and @Shiming94
We have released the SETR benchmark which is based on ViT.
And we will release other Transformer based semantic segmentation benchmarks ASAP.

Hey guy, Thanks soooo much! I will test it on our task. What a great contribution!

wjkim81 pushed a commit to wjkim81/mmsegmentation that referenced this issue Dec 3, 2023
@jin-s13
Add checkpoint and config files (open-mmlab#525)
cf3cf6d 
* add new configs

* add cfgs

* add augmentation configs

* use hasattr to exclude unsupported transforms

* update udp configs
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@Junjun2016 Junjun2016

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
5 participants
@Junjun2016 @jianlong-yuan @MengzhangLI @Shiming94 @clownrat6