Maaz Ullah

Thank @Runist for you comment I have converted yolov4.weights file to .h5 file using Runist [YOLOV4 Repository](https://github.com/Runist/YOLOv4/tree/master) "

I have tried to modify it for multiple ROIs. Maybe I can help you the issue I am facing is speed, its speed decreases as you select more ROIs

I am working on the CPU. wait for1-2 hours I will contact you back

https://github.com/maazullah96/files i have added the files sorry can't send you project because i did not have it now. if you find any issue I will try my best to help...

did your project is working before?

I think it's not a good idea because it slows up things because the first detection takes time and then tracking also takes time in this scenario.

Hi Sorry for the late reply.I have solved this issue Change indices = linear_assignment(cost_matrix) indices = np.hstack([indices[0].reshape(((indices[0].shape[0]), 1)),indices[1].reshape(((indices[0].shape[0]), 1))]) This error is because of previoussklearn library version