search

Cartucho / OpenLabeling

Label images and video for Computer Vision applications
Apache License 2.0
926 stars 265 forks source link

Thickness #14

Closed omrirz closed 6 years ago

omrirz commented 6 years ago

Hi :) [1] fixed the swapped height and width args to yolo_format() [2] added bbox and cross thickness arguments to the argparser.

Cartucho commented 6 years ago

@omrirz Thank you for your help. About the fixed height and width, I believe it was working before. I tested using the data from this tutorial: https://timebutt.github.io/static/how-to-train-yolov2-to-detect-custom-objects/

What do you think? Why do you believe that height and width are in the wrong order?

Cartucho commented 6 years ago

The bbox thickness as an argument is a great idea!

omrirz commented 6 years ago

So it was working fine. Its just that for example x center was calculated with the height argument and the argument passed to height from where this function gets called was actually width and that is why it worked okay :) Just renamed height to width and width to height to match the call of the function.

Cartucho commented 6 years ago

Oh right! Cool!