Open massimiliano96 opened 2 years ago
hi!.. im having a similar issue. using python, opencv and yolov4 tiny, trained with a custom dataset. processing a stream.. a truck is parked... and i get different boxes for each frame .. and the most undesired efect a range of confidences that varies between .7 to .9... and again.. the truck is parket.. not movement in the image... some minor "stream" artifacts
Hi All!
I'm doing object detection with Yolov3 trained on custom data, the problem is that if I Run my network on the same image multiple times I get slightly different result in bounding boxes which differ by a few pixels. Here's my code :
Is this normal? How can I manage this?