search

AlexeyAB / Yolo_mark

GUI for marking bounded boxes of objects in images for training neural network Yolo v3 and v2
https://github.com/AlexeyAB/darknet
The Unlicense
1.8k stars 680 forks source link

negative values in output coordinates left_x, top_y, width, height #184

Open sneha5gsm opened 4 years ago

sneha5gsm commented 4 years ago

Hey everyone, I trained the YOLO network by following all the steps from the link https://github.com/AlexeyAB/darknet#how-to-train-to-detect-your-custom-objects What is the meaning of negative values in the output coordinates that we get when we run the command: darknet.exe detector test cfg/coco.data yolov4.cfg yolov4.weights -ext_output dog.jpg The predictions on my test images were good and the bounding boxes drawn on the prediction.jpg were correct. But some of the values I got were: left_x: 74 top_y: -5 width: 92 height: 167 left_x: -1 top_y: 47 width: 35 height: 119 left_x: 248 top_y: -8 width: 77 height: 159 left_x: 336 top_y: -1 width: 60 height: 144 Is my understanding on the coordinates correct? left_x = top left x coordinate of bounding box top_y = top left y coordinate of bounding box width = with of bounding box height = height of bounding box

Thank you!

kXborg commented 3 years ago

Hi, I got the following link only. Have you found your answer? https://github.com/eriklindernoren/PyTorch-YOLOv3/issues/30