Aditya Mukewar

I am facing the same issue. Please let me know if you have resolved it.

Is topk filter for nms applied before or after the NMS GPU implementation? When I increase the topk, higher confidence bounding boxes appear. Also, total number of objects detected is...

Yes, I am using gen_wts_yoloV5.py from the latest commit.

We are not using any normalization, do you think normalization might fix the issue? Also, is there a way to run the inference using .wts file?

I trained the medium model on the same data and do not see the same level of performance drop. It is not as noisy and the mAP drop is not...

Thanks a lot for the reply. I already tried reducing the frame size to 100. It still kept on running into memory errors. Is there anything else that you think,...

set batch size to 16.