CodeFormer icon indicating copy to clipboard operation
CodeFormer copied to clipboard

[NeurIPS 2022] Towards Robust Blind Face Restoration with Codebook Lookup Transformer

Results 276 CodeFormer issues
Sort by recently updated
recently updated
newest added

Updated crop_align_face.py to use PIL.Image.LANCZOS instead of PIL.Image.ANTIALIAS as it is deprecated and does not work anymore.

When I try to fine-tune codeformer from stage1-stage3 on my own dataset, I found the fine-tuned model is always worse than pre-trained model, any advice? @sczhou

![IMG_3614](https://github.com/user-attachments/assets/f55852cb-43c3-46fc-8ee2-af386aa6042f)

C:\Users\wljkm\CodeFormer\inference_codeformer.py:141: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute...

The body of the person in my picture is also blurred, if only the face is clear, it will feel strange. Is there a way to make the whole picture...