How to obtain the caption of objects?

[Shanzhaguoo]

Hi, excuse me. I wonder how to obtain the caption corresponding to the detected object through the API? Currently, I can only obtain the bbox and label. Hope to receive an answer, thanks! @rentainhe @taras-sereda @leizhangcn @eltociear @JYe9

[xifanii]

@Shanzhaguoo Hi 👋 Thanks for your question!

Currently, if you want to obtain a caption (text description) for each detected object/region, you can do this by calling the Region Vision-Language API:

/v2/task/dinox/region_vl

📘 API usage details and request/response examples can be found in the official documentation here:
https://cloud.deepdataspace.com/en/docs

[Shanzhaguoo]

@xifanii Oh! Thanks for your reply, I'll try it!