miccunifi / ladi-vton

[ACM MM 2023] - LaDI-VTON: Latent Diffusion Textual-Inversion Enhanced Virtual Try-On
Other
412 stars 56 forks source link

How to make inference on only one photo not just the whole dataset? #45

Open ishamstablediffusionapi opened 11 months ago

ishamstablediffusionapi commented 11 months ago

I would like to know how can we make inference on only one single image.

sudip550 commented 10 months ago

I would like to know how can we make inference on only one single image.

Hey.... Do you got an answer??

sjcotto commented 10 months ago

hi, same question

TheFunyBunky commented 9 months ago

Hello, if the image is either from vitonhd or dresscode datasets you can just delete all samples except one and run the inference. If you want to use your custom dataset/image you need to follow steps described in Preprocessing.md to get all poses, masks, etc. Hope someone found it useful.

koluzaeva commented 8 months ago

Hi! image = ["00055_00.jpg", "00013_00.jpg","00176_00.jpg", "00205_00.jpg" ] clothes = ["00782_00.jpg", "00760_00.jpg", "00518_00.jpg", "00705_00.jpg"] df = pd.DataFrame({"image": image, "clothes": clothes}) And save df.to_csv in "test_pairs.txt", index=False, header=False, sep=" "