I have a set of 2D images of lattice structures, which consists of normal images, and images with a broken lattice structure (a crack somewhere).
I am starting a project to predict the images and label the cracks through instance segmentation.
My questions are:
How do you go about creating the labels?
From what I understand, you have to go to a website/app to draw the labels and export them as a JSON file.
If this is correct, which website/app do you guys use?
Upon exporting the JSON file, do you have to do anything to it? Like converting the JSON file to suit the COCO format? If so, how did you guys do it?
