CVPR/GFPGAN-example

1# Installation2 3We now provide a *clean* version of GFPGAN, which does not require customized CUDA extensions. See [here](README.md#installation) for this easier installation.<br>4If you want want to use the original model in our paper, please follow the instructions below.5 61. Clone repo7 8    ```bash9    git clone https://github.com/xinntao/GFPGAN.git10    cd GFPGAN11    ```12 131. Install dependent packages14 15    As StyleGAN2 uses customized PyTorch C++ extensions, you need to **compile them during installation** or **load them just-in-time(JIT)**.16    You can refer to [BasicSR-INSTALL.md](https://github.com/xinntao/BasicSR/blob/master/INSTALL.md) for more details.17 18    **Option 1: Load extensions just-in-time(JIT)** (For those just want to do simple inferences, may have less issues)19 20    ```bash21    # Install basicsr - https://github.com/xinntao/BasicSR22    # We use BasicSR for both training and inference23    pip install basicsr24 25    # Install facexlib - https://github.com/xinntao/facexlib26    # We use face detection and face restoration helper in the facexlib package27    pip install facexlib28 29    pip install -r requirements.txt30    python setup.py develop31 32    # remember to set BASICSR_JIT=True before your running commands33    ```34 35    **Option 2: Compile extensions during installation** (For those need to train/inference for many times)36 37    ```bash38    # Install basicsr - https://github.com/xinntao/BasicSR39    # We use BasicSR for both training and inference40    # Set BASICSR_EXT=True to compile the cuda extensions in the BasicSR - It may take several minutes to compile, please be patient41    # Add -vvv for detailed log prints42    BASICSR_EXT=True pip install basicsr -vvv43 44    # Install facexlib - https://github.com/xinntao/facexlib45    # We use face detection and face restoration helper in the facexlib package46    pip install facexlib47 48    pip install -r requirements.txt49    python setup.py develop50    ```51 52## :zap: Quick Inference53 54Download pre-trained models: [GFPGANv1.pth](https://github.com/TencentARC/GFPGAN/releases/download/v0.1.0/GFPGANv1.pth)55 56```bash57wget https://github.com/TencentARC/GFPGAN/releases/download/v0.1.0/GFPGANv1.pth -P experiments/pretrained_models58```59 60- Option 1: Load extensions just-in-time(JIT)61 62    ```bash63    BASICSR_JIT=True python inference_gfpgan.py --model_path experiments/pretrained_models/GFPGANv1.pth --test_path inputs/whole_imgs --save_root results --arch original --channel 164 65    # for aligned images66    BASICSR_JIT=True python inference_gfpgan.py --model_path experiments/pretrained_models/GFPGANv1.pth --test_path inputs/cropped_faces --save_root results --arch original --channel 1 --aligned67    ```68 69- Option 2: Have successfully compiled extensions during installation70 71    ```bash72    python inference_gfpgan.py --model_path experiments/pretrained_models/GFPGANv1.pth --test_path inputs/whole_imgs --save_root results --arch original --channel 173 74    # for aligned images75    python inference_gfpgan.py --model_path experiments/pretrained_models/GFPGANv1.pth --test_path inputs/cropped_faces --save_root results --arch original --channel 1 --aligned76    ```77