0

I'm a newbie in AI and i'm using labelbox to create my own dataset (instance segmentation) and the annotation output is a single json file.

The issue that i have is that the model that i'm using (Mask RCNN) need to be feed with images with an annotation file in VOC xml for each file.

I need a script that could use the single JSON from labelbox and convert it to multiple images and voc xml annotation file.

Thanks for your help.

Eric Tailleur
  • 1
  • 1

1 Answers1

1

You can use https://app.roboflow.com/ to easily convert them, as per this guide: https://roboflow.com/convert/labelbox-json-to-pascal-voc-xml

Though it realistically wouldn't be all too hard to make a parsing script, at least for bounding boxes, to manually create xml's if 1000 images (robobox's free amount) isn't enough. The Json files contain top left box coordinates as well as width and height. This is basically all the VOC XMLs need.

Matt W
  • 11
  • 1