Skip to content
This repository has been archived by the owner on Apr 16, 2023. It is now read-only.

Do more augmentations #58

Open
gsganden opened this issue Nov 21, 2018 · 2 comments
Open

Do more augmentations #58

gsganden opened this issue Nov 21, 2018 · 2 comments
Labels
cv Computer vision work r&d Research & Development, i.e. a promising but unproven idea

Comments

@gsganden
Copy link
Collaborator

Given that the animals often occupy only a small part of the image, we may get better performance with a lot of augmentations involving high zoom. This kind of approach is often used with medical images, e.g. where a tumor might occupy a small part of the image.

@gsganden gsganden added cv Computer vision work r&d Research & Development, i.e. a promising but unproven idea labels May 25, 2019
@aminzarshenas
Copy link

Hey @gsganden. Is this still a feature you're looking for? I would like to help with that.

@gsganden
Copy link
Collaborator Author

Yes, that would be great. See https://github.com/uptake/autofocus/blob/master/autofocus/train_model/train_multilabel_model.ipynb for a record of the most recent training run.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
cv Computer vision work r&d Research & Development, i.e. a promising but unproven idea
Projects
None yet
Development

No branches or pull requests

2 participants