search

baaivision / CapsFusion

[CVPR 2024] CapsFusion: Rethinking Image-Text Data at Scale
193 stars 5 forks source link

When will the code and dataset available? #4

Open Zhao-Jianing-SUDA opened 11 months ago

Zhao-Jianing-SUDA commented 11 months ago

Great work!

yqy2001 commented 10 months ago

Thank you for your interest in our work. The CapsFus-LLaMA model and distributed inference code have been released, please check it out.

miguelscarv commented 9 months ago

What about the 10M version of the dataset?

rahimentezari commented 9 months ago

Do you plan to release 100M version?

yqy2001 commented 9 months ago

Do you plan to release 100M version?

@miguelscarv @rahimentezari Hello, we have released the CapsFusion-120M dataset, please check it out!

yqy2001 commented 9 months ago

What about the 10M version of the dataset?

@miguelscarv The 10M version of the dataset might take some more time to be released, as there are some complications involved with releasing images. Not sure whether the current 120M format (url + captions) could satisfy your need?