Hi everyone, I got confused about the inpainting model input format.
I think it is different from the stablediffusion inpainting. I suppose it takes a (H,W,3) image as a condition and set the masked area as -1 as the control. I tried this when I finetune the inpainting model. However, it cannot work. I attached my code and hope that someone can let me know if I did anything wrong. Thanks.
Hi everyone, I got confused about the inpainting model input format.
I think it is different from the stablediffusion inpainting. I suppose it takes a (H,W,3) image as a condition and set the masked area as -1 as the control. I tried this when I finetune the inpainting model. However, it cannot work. I attached my code and hope that someone can let me know if I did anything wrong. Thanks.