inst-inpaint
inst-inpaint copied to clipboard
Any plans to do some research on masking based removing methods?
It's a great work! I wonder if this method can be transferred to mask based removing? As instruction sometimes cannot convey the user's accurate intention! I have tried SD-Inpaint, but its removing ability is not that good, it usually generate some unwanted objects. Do you have some ideas about diffusion based masking removing strategy?
It's a great work! I wonder if this method can be transferred to mask based removing? As instruction sometimes cannot convey the user's accurate intention! I have tried SD-Inpaint, but its removing ability is not that good, it usually generate some unwanted objects. Do you have some ideas about diffusion based masking removing strategy?
Actually, in this version, how can you specify which object is desired to be removed? In which parameter should we explain (prompt) the script about the object that needs to be removed?
It's a great work! I wonder if this method can be transferred to mask based removing? As instruction sometimes cannot convey the user's accurate intention! I have tried SD-Inpaint, but its removing ability is not that good, it usually generate some unwanted objects. Do you have some ideas about diffusion based masking removing strategy?
@lawsonxwl I have the same concentration as you did , and even the latest model seems to fail on the simple object removal ,using mask as guided. I have tried sd-inpaint and powerpaint_v2 . Always worsen than even lama. Did you find any solutions?
We have been working on this issue recently. Today we published the preprint of our paper, you can check it at: https://yigitekin.github.io/CLIPAway/. Our method adapts Stable Diffusion to the object removal task. The code will be released soon, stay tuned!