Open qiqigit opened 4 months ago
Thank you so much for sharing this great work!
I noticed that the clothing-agnostic masks automatically generated on huggingface are slightly different from those provided by the original VITON-HD dataset. Could you please specify how exactly the masks are generated on huggingface?
e.g. mask provided by the VITON-HD dataset:
mask generated on the Huggingface demo:
Mask is being generated using pose detection in Vton
Thank you so much for sharing this great work!
I noticed that the clothing-agnostic masks automatically generated on huggingface are slightly different from those provided by the original VITON-HD dataset. Could you please specify how exactly the masks are generated on huggingface?
e.g. mask provided by the VITON-HD dataset:
mask generated on the Huggingface demo: