-
Notifications
You must be signed in to change notification settings - Fork 13
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
Already trained model. #9
Comments
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. goturn-pytorch/src/scripts/demo_folder.py Line 213 in bb3b3b4
How to test: This will guide as to how to test for new video sample |
Yes exactly, my network gives the bounding box coordinates. I am aiming to use those coordinates to initiate goturn tracker. |
@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? |
Since I'm using Pytorch-Lightning, there are few changes to it in the new version. |
I got segmentation fault when I press """s""" for tracking. |
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?The text was updated successfully, but these errors were encountered: