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

Already trained model. #9

Open
berkantay opened this issue Nov 16, 2020 · 5 comments
Open

Already trained model. #9

berkantay opened this issue Nov 16, 2020 · 5 comments

Comments

@berkantay
Copy link

Hello @nrupatunga i am planning to use goturn tracker with yolov5. I have my trained weights and detection working well. Now I want to integrate goturn tracker to my code. I have my_custom_model.pt. Can I use this for tracking?

@nrupatunga
Copy link
Owner

yes, so if I understand correctly, your object detection model gives an initial bounding box for the tracking to continue,

so you can refer to the following test script(might need some modifications) and can set your initial bounding box.

if key == ord('s'):

How to test: This will guide as to how to test for new video sample
https://github.com/nrupatunga/goturn-pytorch#testing

@berkantay
Copy link
Author

Yes exactly, my network gives the bounding box coordinates. I am aiming to use those coordinates to initiate goturn tracker.

@berkantay
Copy link
Author

@nrupatunga I have one more question. Yolov5 framework that I use for detection uses newer version of torch, pillow etc. would it matter if I mind to use those versions for your code?

@nrupatunga
Copy link
Owner

Since I'm using Pytorch-Lightning, there are few changes to it in the new version.
Let me know if you face any issues, would look into it.

@berkantay
Copy link
Author

I got segmentation fault when I press """s""" for tracking.

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