CodeFormer
CodeFormer copied to clipboard
[NeurIPS 2022] Towards Robust Blind Face Restoration with Codebook Lookup Transformer
When I load an image into a script, I occasionally get spots on the photo. With what it can be connected? 
Which linux repo contains the following packages? addict, future, lmdb, numpy, opencv-python, Pillow, pyyaml, requests, scikit-image, scipy, tb-nightly, torch>=1.7.1, torchvision, tqdm, yapf, lpips, gdown # supports downloading the large file...
Hi.Photos sometimes after processing the lip color will become blue-gray, if I change the background color of the photo lip color right normal.  
I use V100 GPU and batch is 1 , the loss and network same as paper , and my code reference GPEN training code .
**Pls howto fix memory error?** "Error CUDA out of memory. Tried to allocate 930.00 MiB (GPU 0; 3.82 GiB total capacity; 849.56 MiB already allocated; 938.75 MiB free; 1.66 GiB...
I installed it locally, and checked the similar hugging face page: [https://huggingface.co/spaces/sczhou/CodeFormer](url) but there are a few issues, maybe I'm just not getting it. 1. I don't see how to...
  In most of cases eyes are green while most of Asians having dark eyes. Also I suggest to expand detected face box to cover hairs too.
Hi, can you release your inpainting training code ?