Lama-cleaner

Lama Cleaner is no living on 🤗 Huggingface Spaces! https://huggingface.co/spaces/Sanster/Lama-Cleaner-lama

<img width="1760" alt="image" src="https://user-images.githubusercontent.com/3998421/202192760-a4f783f9-da39-4fac-a38d-2cb117a5e402.png">

- Add upload mask button. Once uploaded, you can rerun inpainting with the current mask by clicking the play button. The mask must have the same shape as the image.
<img width="208" alt="image" src="https://user-images.githubusercontent.com/3998421/202080438-dbd441e3-c1d3-4811-a9a3-81d4d34fbefe.png">
mask example(white area is mask area):

- Better error message when CUDA OOM
- Add `--disable-model-switch` when start server. Disable model switch from frontend
- sd1.5
- Faster model loading thanks to [diffusers 0.7.0 update](https://github.com/huggingface/diffusers/releases/tag/v0.7.0):
- Negative Prompt
- Add `k_euler` and `k_euler_a` sampler
- Make the cropper easier to drag and resize: https://github.com/Sanster/lama-cleaner/pull/111

- remove `torch.amp.autocast` in sd1.5, run 25% faster(tested on 3080)
- remove `Strength` from the sd side panel, it has nothing to do with inpainting
- fix the flaskwebgui version

win_one_click_installer
- Windows One Click Installer: Follow the guide here to install: https://github.com/Sanster/lama-cleaner/blob/main/scripts/README.md
- Add Buy Me a Coffee btn on page



one_click_installer
Follow the guide here to install it: https://github.com/Sanster/lama-cleaner/blob/main/scripts/README.md

fix sd1.5 inpainting size bug

- Add stable-diffusion-inpainting model(sd.5) from https://github.com/runwayml/stable-diffusion. Compared to sd1.4 it has better results
- new example video 🎉

https://user-images.githubusercontent.com/3998421/196976498-ba1ad3ab-fa18-4c55-965f-5c6683141375.mp4

1. Bug Fix: correctly pass cv2 config to the server
2. Feature: press alt/opt and move the mouse left/right to change brush size


https://user-images.githubusercontent.com/3998421/194760412-581eaa69-12bf-4f33-b8fa-d89d131d1a23.mov