xiezhy6 / GP-VTON

Official Implementation for CVPR2023 paper "GP-VTON: Towards General Purpose Virtual Try-on via Collaborative Local-Flow Global-Parsing Learning"
475 stars 83 forks source link

person_clothes_ratio_train.txt #33

Open michael312319999 opened 1 year ago

michael312319999 commented 1 year ago

Does anyone know how the ratios are computed in person_clothes_ratio_train?

michael312319999 commented 1 year ago

What happens if I don't include it in the training? Will the results be worse?

Vinbert0203 commented 1 year ago

I have the same issue

michael312319999 commented 1 year ago

Is it personal image (width/height) on a cloth image (width/height) or something similar ?

liupengcnu commented 11 months ago

Does anyone know how the ratios are computed in person_clothes_ratio_train?

have you sove this problem?