search

cuiaiyu / dressing-in-order

(ICCV'21) Official code of "Dressing in Order: Recurrent Person Image Generation for Pose Transfer, Virtual Try-on and Outfit Editing" by Aiyu Cui, Daniel McKee and Svetlana Lazebnik
https://cuiaiyu.github.io/dressing-in-order
Other
507 stars 126 forks source link

Pose Transfer output image is empty #61

Closed onchiptech closed 1 year ago

onchiptech commented 1 year ago

Hi, When I am running the demo I get an empty output image for pose transfer and other tasks. What could be wrong?

load vgg ckpt from torchvision dict. [init] init pre-trained model vgg. initialize network with orthogonal initialize network with orthogonal load ckpt from /content/dior/DIOR_64/latest_net_Flow.pth. [init] init pre-trained model flownet. initialize network with orthogonal [init] frozen net netVGG. [init] frozen net netFlow. [init] frozen net netE_attr. [init] frozen net netE_attr. not exsits /content/dior/checkpoints/DIORv1_64/latest_net_E_attr.pth not exsits /content/dior/checkpoints/DIORv1_64/latest_net_G.pth not exsits /content/dior/checkpoints/DIORv1_64/latest_net_VGG.pth not exsits /content/dior/checkpoints/DIORv1_64/latest_net_Flow.pth ---------- Networks initialized ------------- [Network E_attr] Total number of parameters : 1.191 M [Network G] Total number of parameters : 16.501 M [Network VGG] Total number of parameters : 0.113 M [Network Flow] Total number of parameters : 6.608 M

torch.Size([18, 256, 176]) WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).

out

onchiptech commented 1 year ago

After copying the model file to checkpoints it is working! Now I get the output image.