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

VITON-HD is missing cloth-parse data for training dataset #27

Open ybchen97 opened 1 year ago

ybchen97 commented 1 year ago

Hi @xiezhy6, thanks for authoring this wonderful work!

I realized that the cloth_parse data is missing from the training dataset for VITON-HD-512:

image

Without this we can't perform any training on the VITON-HD-512 dataset with your method. Could you please upload the cloth_parse data? That would be very much appreciated.

Thank you.

求大佬上传 VITON-HD-512 cloth_parse 数据集, 万分感谢

ChengBinJin commented 9 months ago

@ybchen97 You can resize from other resolutions, for example resizing from 1024 or 256.