Currently, I am doing a semantic segmentation task by using mask rcnn with a small dataset and trying to aug more images by this repo. However, at this stage, I have understood how to import the LabelMe JSON file and do corresponding augmentations that I want, but I don't know how to export the results of augmented polygon back to JSON format. How can I make it? or do you use this repo before or during the training process such that you don't need any exportation, right?
Hope to see how you use this repo in certain computer vision task, thanks.
Hi,
Currently, I am doing a semantic segmentation task by using mask rcnn with a small dataset and trying to aug more images by this repo. However, at this stage, I have understood how to import the LabelMe JSON file and do corresponding augmentations that I want, but I don't know how to export the results of augmented polygon back to JSON format. How can I make it? or do you use this repo before or during the training process such that you don't need any exportation, right?
Hope to see how you use this repo in certain computer vision task, thanks.
JX