<confidence> <left> <top> <right> <bottom>

Open shahrukh131 opened this issue 6 years ago • 2 comments

how can i find from predicted box's from the image?

Mar 05 '19 17:03 shahrukh131

Hello, you need to run an object detector over the images, the detector will give you 1 bounding box per object.

Which detector are you using?

Mar 19 '19 09:03 Cartucho

I have the groundtruths and detections in the form of x_min,y_min,x_max,y_max. Can I use the order x_min,y_max,x_max,y_min to match the corresponding order of <left> <top> <right> <bottom> mentioned in the txt file or should I change it something else?

Jun 19 '19 08:06 jiteshm17