Where the image and captions come from?

Question

Where the image and captions come from?

linjing7 opened this issue a year ago · 5 comments

Hi, thanks for your great work! I have a question about the data generation:
Could you please tell me which datasets the images and captions come from?

Answer 1 · 2023-12-19T09:44:22.000Z

We based on COCO dataset

Answer 2 · 2023-12-19T09:45:44.000Z

Okay, thanks for your kind reply!

Answer 3 · 2023-12-19T14:44:48.000Z

Hi, there are five captions for each image in COCO, do you use all the captions or randomly choose one?

Answer 4 · 2023-12-20T02:08:56.000Z

We use 5 captions, or you can use the llava-instruction dataset, which generates dense captions. linjing7 ***@***.***> 于2023年12月19日周二 22:44写道：

…

Hi, there are five captions for each image in COCO, do you use all the captions or randomly choose one? — Reply to this email directly, view it on GitHub <#25 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARTC2EHKBTA7WB6YK7LO6N3YKGR6XAVCNFSM6AAAAABA22TRLWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNRSHA4TCNBTGY> . You are receiving this because you commented.Message ID: ***@***.***>

Answer 5 · 2023-12-20T03:38:15.000Z

Okay, thanks for your reply.