pose-with-style
pose-with-style copied to clipboard
artifacts in outputs
Hi: This project is very interesting :) But when I tried to run reposing with default samples and pretrained models, its' output looks not very good, jagged contours and artifacts. I tried to repose a frontal body with target frontal body, still failed. My environment: pytorch 1.9.0, torchvision 0.10.0, torchaudio 0.9.0, cudatoolkit 10.2.89 and opencv-python(pip). 1080Ti
input_name fashionWOMENSkirtsid0000177102_1front
target_name fashionWOMENBlouses_Shirtsid0000635004_1front
input_name fashionWOMENBlouses_Shirtsid0000635004_1front
target_name fashionWOMENSkirtsid0000177102_1front
Both pretrained models and pretrained(sarkar) failed. Did I miss something?
Well, download the official result from the project page, and go for the adventure to the land of horror :D