Using all things/stuffs for inference

[ffahmed]

I want to do inference for all possible things/stuffs XDecoder is trained on.
How do I construct the metadatacatalog (that would be used in inference), contain all these classes?

[MaureenZOU]

Please refer to here:

X-Decoder/demo/demo_panoseg.py

Line 56 in f3a1043

thing_classes = ['car','person','traffic light', 'truck', 'motorcycle']

[ffahmed]

I meant to say if we don't know what things and stuffs we want from a picture, I want to tell it to give me all possible things and stuffs in panoptic seg. How do I do it?