sangyun884 / HR-VITON

Official PyTorch implementation for the paper High-Resolution Virtual Try-On with Misalignment and Occlusion-Handled Conditions (ECCV 2022).
825 stars 171 forks source link

How to get DensePose images #81

Open ckalla-otto opened 10 months ago

ckalla-otto commented 10 months ago

Hey, can someone give me the command to get the correct DensePose images using Detectron2 used in this implementation? Currently I use this command, which gives me the wrong images:

python apply_net.py show configs/densepose_rcnn_R_50_FPN_s1x.yaml densepose_rcnn_R_50_FPN_s1x.pkl $resized_filename dp_segm --output image_densepose_segm.png