search

yisol / IDM-VTON

[ECCV2024] IDM-VTON : Improving Diffusion Models for Authentic Virtual Try-on in the Wild
https://idm-vton.github.io/
3.64k stars 579 forks source link

Could I use my own pictures？ #23

Open Yuziyue opened 4 months ago

Yuziyue commented 4 months ago

Nice job! I have used inference.py to test VITON-HD dataset. However, I wonder if the codes support to use my own pictures to generate images ( Instead of by using demo 🙂)?

rokonezoko55555555 commented 4 months ago

https://huggingface.co/spaces/yisol/IDM-VTON you can also use your own img ,just tap on on upload button 64d1f417-a7e2-4945-b77e-3d9feff167f4 0afc7c26-1a20-449b-841d-e16086a3cdf0

Yuziyue commented 4 months ago

https://huggingface.co/spaces/yisol/IDM-VTON you can also use your own img ,just tap on on upload button 64d1f417-a7e2-4945-b77e-3d9feff167f4 0afc7c26-1a20-449b-841d-e16086a3cdf0

THX😀，but what I want is using codes rather than a html demo, since I have successfully run the inference code. 😂

yisol commented 4 months ago

Hello, you can run inference.py with your own dataset for batch generation. You have to prepare mask image and densepose image. You can refer this link for preparing densepose image.

Yuziyue commented 4 months ago

Hello, you can run inference.py with your own dataset for batch generation. You have to prepare mask image and densepose image. You can refer this link for preparing densepose image.

Very thanks，now I have followed this link and got (agnostic-v3.2) images like 14679_00 but I find what the inference code needs is (agnostic-mask) images like (I'm not sure🙂) 14679_00_mask The link doesn't show how to get this, could you help me?

gillbates commented 4 months ago

same question here that: how could we get these black white mask images ...