How to make a panoramic segmentation dataset that contains both thing and stuff?

[canglangzhige]

Thanks for your work!
The script detection2panoptic_coco_format.py converts detection COCO format to panoptic COCO format.
But the detection COCO format only includes things, and the panoptic segmentation dataset also requires stuff. How to make a panoptic segmentation dataset that contains both thing and stuff?

[SumanthReddyKaliki]

I have same question as well. I have my dataset masked in both instance and semantic format. I would like to combine them to create a panoptic dataset for training. Could someone please let me know how this should be done. Thank you.