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

Wrong height and width in some annotations #6

Open
fei-chang opened this issue Mar 17, 2023 · 1 comment
Open

Wrong height and width in some annotations #6

fei-chang opened this issue Mar 17, 2023 · 1 comment

Comments

@fei-chang
Copy link

@xiaojie1017 Thanks for building and open-sourcing this dataset!

However, I noticed that the height and the width are wrongly annotated in some xml files.

To verify such errors exist, you could refer to the xml annotations of the two images below.

  • 01e070bf7ddddf7afca0d4556418f7bdaa841a29.jpg
  • 0cde37c020e9bfb176f0748ecb10107fbfdf9047.jpg

The similiar error exists in around 10% or more of this dataset. It cause problems when the width and height are directly read from the xml annotation instead of the image file (I found this problem when I try to convert these annotations into yolo format).

A revision on the annotation would be nice, or there should be a note in the readme suggesting that the height and width in xml files are not trustworthy.

@xiaojieli0903
Copy link
Contributor

@fei-chang Thank you for bringing this issue to our attention and for your valuable feedback on the dataset. I apologize for the inconvenience caused by the incorrect annotations for height and width in some XML files. I am committed to fixing these errors promptly.

To address this, I will include a Python script in the dataset repository that automatically generates the correct width and height for each image, ensuring that the data can be used reliably without manual corrections. This script will help maintain the integrity of the dataset and facilitate its use in various formats, including YOLO.

If it's possible, could you also share the script you used to convert the annotations to YOLO format? It would be immensely helpful for ensuring that our dataset can be seamlessly integrated into different workflows.

Thank you once again for your contribution, and I look forward to resolving this issue quickly.

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