Testing needs mask and pose imformation along with model image and target cloth?

benquick123 / C-VTON

C-VTON: Context-Driven Image-Based Virtual Try-On Network

Eclipse Public License 2.0

142 stars 31 forks source link

Testing needs mask and pose imformation along with model image and target cloth? #1

Closed hamaadtahiir closed 3 years ago

hamaadtahiir commented 3 years ago

I have done the testing with pretrained model on vton dataset. It looks to me that the testing needs all the other information along with model, cloth and edge, such as densepose, masks etc. Shouldnt testing only need model image, cloth and its edge?

benquick123 commented 3 years ago

I see why there might be some confusion. The dataset that loads the data are the same no matter if it is used for training or testing. However, while you will have to tweak the VitonDataset.py and MPVDataset.py to not load additional assets during testing, only the images you mentioned are used in the testing phase.

hamaadtahiir commented 2 years ago

so it means that the segmentation information is required in the testing phase