Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[Feature] The performance of transformer-based rotation object detection is not as good as ResNet50. #1058

Open
ZhenboZhao77 opened this issue Aug 17, 2024 · 0 comments

Comments

@ZhenboZhao77
Copy link

What's the feature?

Recently, I conducted experiments comparing different models and found that using transformer-based models for rotation object detection did not perform well. I recall reading that the QKV mechanism might not be well-suited for objects with large scale variations. Could anyone share more detailed papers or explanations on the disadvantages of QKV in the application of rotation object detection?

Any other context?

No response

# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant