Open cian0 opened 1 year ago
Thanks. Can you give me a link to what you said? That's possible if we train the model with higher resolution.
https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features
I don't think there's a need to train with a higher resolution for this as it's simply a technique to resize and repaint the masked area. This is usually used for faces in SD
Under inpaint area, the doc says:
Inpaint area Normally, inpainting resizes the image to the target resolution specified in the UI. With Inpaint area: Only masked enabled, only the masked region is resized, and after processing it is pasted back to the original picture. This allows you to work with large pictures and render the inpainted object at a much larger resolution.
I understand it now. This would be a good feature. We can only edit a part of the image in high resolution.
Yup though I'm not sure if I get why outputs would have blurred faces as per documentation..
Here are some images from the SAM dataset. You can find more at https://segment-anything.com/dataset/index.html We are solving this problem by finetuning models on the Laion dataset.
Similar to high res inpainting in SD webui, would this be possible?
I want to add this feature to my demo application. But I couldn't find the codes. Can you help me? Demo: https://huggingface.co/spaces/ArtGAN/Stable-Diffusion-ControlNet-WebUI
Similar to high res inpainting in SD webui, would this be possible?
I want to add this feature to my demo application. But I couldn't find the codes. Can you help me? Demo: https://huggingface.co/spaces/ArtGAN/Stable-Diffusion-ControlNet-WebUI
I believe it's here :
Similar to high res inpainting in SD webui, would this be possible?