search

BrandonHanx / mmf

[ECCV 2022] FashionViL: Fashion-Focused V+L Representation Learning
https://mmf.sh/
Other
57 stars 7 forks source link

question for fashion-gen dataset #16

Closed houjie8888 closed 1 year ago

houjie8888 commented 1 year ago

I was wondering what resolution of the FashionGen dataset you used, as I noticed it was in PNG format. The version I had access to was only 256 resolution. If you used the high-resolution version mentioned in the FashionGen paper, could you please let me know how you obtained it? I was unable to reach the original author at that time. Thank you very much for your help!

BrandonHanx commented 1 year ago

For fashiongen, we used 256 png provided by the authors.

houjie8888 commented 1 year ago

thanks.