Zongsheng
@OAOAOAOAOAOooo
its called InvSR and is a diffusion based upscale method which only requires one to five steps code: github.com/zsyOAOA/InvSR demo: huggingface.co/spaces/OAOA/In…
Our new work InvSR: Image super-resolution via Diffusion Inversion. Welcome to have a try. @ccloy @KangLiao929 Project page: github.com/zsyOAOA/InvSR Hugging Face: huggingface.co/spaces/OAOA/In… Replicate: replicate.com/zsyoaoa/invsr
Our work on image super-resolution has been accepted by NeurIPS 2023 as a spotlight. It achieves impressive SR results only with 15 diffusion steps. Welcome to try our demo. @_akhaliq @ccloy @Iceclearwjy Demo: replicate.com/cjwbw/resshift Project page: zsyoaoa.github.io/projects/ressh…
Our work devise a beautiful way to leverage the diffusion prior in the task of blind face restoration. The error contraction conducted by diffusion model makes our method very robustness to severe degradations.
DifFace: Blind Face Restoration with Diffused Error Contraction abs: arxiv.org/abs/2212.06512 github: github.com/zsyOAOA/DifFace
DifFace: Blind Face Restoration with Diffused Error Contraction abs: arxiv.org/abs/2212.06512 github: github.com/zsyOAOA/DifFace