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

Feature Request: Ability to Review and Clean Image Patches by Label #25

Open
m-h-williams opened this issue Sep 16, 2024 · 4 comments
Open
Assignees
Labels
good first issue Good for newcomers

Comments

@m-h-williams
Copy link

@Jordan-Pierce I’m really excited to see all the improvements you’ve been making to CoralNet-toolbox! I also noticed the change of the working branch to dev—great move.

The addition of the link to information about the hyperparameters in the training model window is incredibly thoughtful. It's a helpful resource when making adjustments. The adjustments to the dataset creation window are also very useful, especially for understanding which labels have sufficient annotations to be included in training, and which ones need more data.

There’s no rush on our end, but I wanted to check in and see if you’ve had a chance to think more about adding the ability to clean up image patches via the image patch folders. I know the patch extractor was an older tool, so image patch folders aren’t created by default in the new toolbox, except when datasets are generated. I think it could be really helpful to have that capability as a part of the new workflow.

I’d be happy to set up a time to chat about this in more detail if you’d like!

@m-h-williams m-h-williams changed the title Feature request: Ability to regenerate patches csv file after deleting images patches in discrete label folders Feature Request: Ability to Review and Clean Image Patches by Label Sep 16, 2024
@Jordan-Pierce
Copy link
Owner

Hi @m-h-williams still working on some other stuff (I wanted to finish some foundational stuff, such as polygons) before moving on to the table. It makes things a bit easier later down the line.

@m-h-williams
Copy link
Author

Hi @Jordan-Pierce,

Thanks for the update! I completely understand, and I really appreciate all the foundational work you’re doing, especially with the polygons. It sounds like it’ll make things more streamlined and efficient down the road.

No rush on our end—we’re excited to see how everything evolves! Let me know if there’s anything I can do to help, and I’d be happy to chat more when the time is right.

@Jordan-Pierce
Copy link
Owner

Hey @m-h-williams I've recently finished all of what I really wanted to get done for the time being, so I'll be starting on this next. Cheers.

@m-h-williams
Copy link
Author

@Jordan-Pierce I've been experimenting with the SAM tool using our imagery, and it’s performing really well on various algae species and the different substrates we encounter (i.e., boulder, silt, and cobble). I’m really impressed with the new capabilities and am excited to see how this next stage develops! If there’s anything I can do to assist, please feel free to reach out.

@Jordan-Pierce Jordan-Pierce added the good first issue Good for newcomers label Dec 19, 2024
@Jordan-Pierce Jordan-Pierce self-assigned this Dec 19, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
good first issue Good for newcomers
Projects
None yet
Development

No branches or pull requests

2 participants