search

microsoft / unilm

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities
https://aka.ms/GeneralAI
MIT License
19.08k stars 2.43k forks source link

[Kosmos-2] Bounding Box Format #1560

Open jujeongho0 opened 1 month ago

jujeongho0 commented 1 month ago

Thanks for the Great Work!!!

Kosmos-2 uses special tokens like to represent bounding boxes, rather than coordinates like (x1, y1, x2, y2).

Can I know the location of the code that turns coordinates into special tokens?

Thank you.

pengzhiliang commented 1 week ago

Hello, @jujeongho0 You can check it in this file .