image enhanced api developed

.gitignore

@@ -0,0 +1,143 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+# Usually these files are written by a python script from a template
+# before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+*.po
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+# Pipfile.lock
+
+# poetry
+# Poetry explicitly recommends committing the poetry.lock file
+# poetry.lock
+
+# pdm
+# According to pdm-project/pdm#368, it is recommended to include pdm.lock in version control.
+# pdm.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static analysis results
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+api_inputs/
+api_outputs/

README.md

@@ -0,0 +1,95 @@
+# Image Enhancer
+
+High-resolution image enhancement powered by AI.
+
+## Before / After Example
+
+| Before                                                           | After                                                                                                 |
+| ---------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------- |
+| ![Before](api_inputs/input_5645f8f6b3fc470d884486625205ee53.jpg) | ![After](api_outputs/5645f8f6b3fc470d884486625205ee53/input_5645f8f6b3fc470d884486625205ee53_out.jpg) |
+
+## Description
+
+This application provides a simple web interface to enhance images using an AI upscaling model. Upload your low-resolution images and get high-resolution results.
+
+## Features
+
+- Web UI for easy image uploading and enhancement.
+- API endpoint for programmatic access (`/enhance`).
+- Selectable upscaling models and scale factors.
+- Optional face enhancement.
+- Input and output images are saved in `api_inputs` and `api_outputs` respectively.
+
+## Installation
+
+1.  **Clone the repository:**
+    ```bash
+    git clone https://huggingface.co/sayed99/Image-Enhancer
+    cd Image-Enhancer
+    ```
+2.  **Create Conda Environment:**
+    Set up the necessary environment using the provided `environment.yml` file:
+
+    ```bash
+    conda env create -f environment.yml
+    conda activate esrgan-env
+    ```
+
+    _(Note: The environment name is defined within the `environment.yml` file)_
+
+3.  **Download Model Weights:**
+    The required model weights (`.pth` files) need to be placed in the `Real-ESRGAN/weights/` directory. Common models include:
+    - `RealESRGAN_x4plus.pth`
+    - `RealESRGAN_x2plus.pth`
+      You can usually find these linked from the original Real-ESRGAN repository or other model sources.
+
+## Usage
+
+Run the application using the provided script:
+
+```bash
+python run.py
+```
+
+This will:
+
+1.  Start the backend API server (usually on `http://localhost:8000`).
+2.  Start the Streamlit web interface (usually on `http://localhost:8501`).
+3.  Open the web interface in your default browser.
+
+Navigate to the web interface, upload an image, select your desired options (model, scale, face enhancement), and click "Enhance Image".
+
+## API Usage
+
+You can also interact with the API directly.
+
+**Enhance Endpoint:** `POST /enhance/`
+
+**Form Data:**
+
+- `file`: The image file to upload.
+- `model_name` (optional, default: `RealESRGAN_x4plus`): Model to use (e.g., `RealESRGAN_x4plus`, `RealESRGAN_x2plus`).
+- `outscale` (optional, default: `4.0`): The desired output scale factor (e.g., `2.0`, `4.0`).
+- `face_enhance` (optional, default: `false`): Boolean flag to enable face enhancement.
+- `fp32` (optional, default: `false`): Boolean flag to use FP32 precision.
+
+**Example using `curl`:**
+
+```bash
+curl -X POST "http://localhost:8000/enhance/" \
+     -F "file=@/path/to/your/image.jpg" \
+     -F "model_name=RealESRGAN_x4plus" \
+     -F "outscale=4.0" \
+     -o enhanced_image.jpg
+```
+
+## Notes
+
+- Ensure the API server is running before using the Streamlit app or sending direct API requests.
+- The application uses significant resources (RAM, potentially GPU if configured).
+- Input images are saved in the `api_inputs` directory.
+- Output images are saved in subdirectories within the `api_outputs` directory, named by a unique request ID.
+
+---
+
+Powered by Real-ESRGAN

Real-ESRGAN/gfpgan/weights/detection_Resnet50_Final.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d1de9c2944f2ccddca5f5e010ea5ae64a39845a86311af6fdf30841b0a5a16d
+size 109497761

Real-ESRGAN/gfpgan/weights/parsing_parsenet.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d558d8d0e42c20224f13cf5a29c79eba2d59913419f945545d8cf7b72920de2
+size 85331193

Real-ESRGAN/inference_realesrgan.py

@@ -0,0 +1,166 @@
+import argparse
+import cv2
+import glob
+import os
+from basicsr.archs.rrdbnet_arch import RRDBNet
+from basicsr.utils.download_util import load_file_from_url
+
+from realesrgan import RealESRGANer
+from realesrgan.archs.srvgg_arch import SRVGGNetCompact
+
+
+def main():
+    """Inference demo for Real-ESRGAN.
+    """
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-i', '--input', type=str, default='inputs', help='Input image or folder')
+    parser.add_argument(
+        '-n',
+        '--model_name',
+        type=str,
+        default='RealESRGAN_x4plus',
+        help=('Model names: RealESRGAN_x4plus | RealESRNet_x4plus | RealESRGAN_x4plus_anime_6B | RealESRGAN_x2plus | '
+              'realesr-animevideov3 | realesr-general-x4v3'))
+    parser.add_argument('-o', '--output', type=str, default='results', help='Output folder')
+    parser.add_argument(
+        '-dn',
+        '--denoise_strength',
+        type=float,
+        default=0.5,
+        help=('Denoise strength. 0 for weak denoise (keep noise), 1 for strong denoise ability. '
+              'Only used for the realesr-general-x4v3 model'))
+    parser.add_argument('-s', '--outscale', type=float, default=4, help='The final upsampling scale of the image')
+    parser.add_argument(
+        '--model_path', type=str, default=None, help='[Option] Model path. Usually, you do not need to specify it')
+    parser.add_argument('--suffix', type=str, default='out', help='Suffix of the restored image')
+    parser.add_argument('-t', '--tile', type=int, default=0, help='Tile size, 0 for no tile during testing')
+    parser.add_argument('--tile_pad', type=int, default=10, help='Tile padding')
+    parser.add_argument('--pre_pad', type=int, default=0, help='Pre padding size at each border')
+    parser.add_argument('--face_enhance', action='store_true', help='Use GFPGAN to enhance face')
+    parser.add_argument(
+        '--fp32', action='store_true', help='Use fp32 precision during inference. Default: fp16 (half precision).')
+    parser.add_argument(
+        '--alpha_upsampler',
+        type=str,
+        default='realesrgan',
+        help='The upsampler for the alpha channels. Options: realesrgan | bicubic')
+    parser.add_argument(
+        '--ext',
+        type=str,
+        default='auto',
+        help='Image extension. Options: auto | jpg | png, auto means using the same extension as inputs')
+    parser.add_argument(
+        '-g', '--gpu-id', type=int, default=None, help='gpu device to use (default=None) can be 0,1,2 for multi-gpu')
+
+    args = parser.parse_args()
+
+    # determine models according to model names
+    args.model_name = args.model_name.split('.')[0]
+    if args.model_name == 'RealESRGAN_x4plus':  # x4 RRDBNet model
+        model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=4)
+        netscale = 4
+        file_url = ['https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth']
+    elif args.model_name == 'RealESRNet_x4plus':  # x4 RRDBNet model
+        model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=4)
+        netscale = 4
+        file_url = ['https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.1/RealESRNet_x4plus.pth']
+    elif args.model_name == 'RealESRGAN_x4plus_anime_6B':  # x4 RRDBNet model with 6 blocks
+        model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=6, num_grow_ch=32, scale=4)
+        netscale = 4
+        file_url = ['https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth']
+    elif args.model_name == 'RealESRGAN_x2plus':  # x2 RRDBNet model
+        model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=2)
+        netscale = 2
+        file_url = ['https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth']
+    elif args.model_name == 'realesr-animevideov3':  # x4 VGG-style model (XS size)
+        model = SRVGGNetCompact(num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=16, upscale=4, act_type='prelu')
+        netscale = 4
+        file_url = ['https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-animevideov3.pth']
+    elif args.model_name == 'realesr-general-x4v3':  # x4 VGG-style model (S size)
+        model = SRVGGNetCompact(num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=32, upscale=4, act_type='prelu')
+        netscale = 4
+        file_url = [
+            'https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-wdn-x4v3.pth',
+            'https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth'
+        ]
+
+    # determine model paths
+    if args.model_path is not None:
+        model_path = args.model_path
+    else:
+        model_path = os.path.join('weights', args.model_name + '.pth')
+        if not os.path.isfile(model_path):
+            ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
+            for url in file_url:
+                # model_path will be updated
+                model_path = load_file_from_url(
+                    url=url, model_dir=os.path.join(ROOT_DIR, 'weights'), progress=True, file_name=None)
+
+    # use dni to control the denoise strength
+    dni_weight = None
+    if args.model_name == 'realesr-general-x4v3' and args.denoise_strength != 1:
+        wdn_model_path = model_path.replace('realesr-general-x4v3', 'realesr-general-wdn-x4v3')
+        model_path = [model_path, wdn_model_path]
+        dni_weight = [args.denoise_strength, 1 - args.denoise_strength]
+
+    # restorer
+    upsampler = RealESRGANer(
+        scale=netscale,
+        model_path=model_path,
+        dni_weight=dni_weight,
+        model=model,
+        tile=args.tile,
+        tile_pad=args.tile_pad,
+        pre_pad=args.pre_pad,
+        half=not args.fp32,
+        gpu_id=args.gpu_id)
+
+    if args.face_enhance:  # Use GFPGAN for face enhancement
+        from gfpgan import GFPGANer
+        face_enhancer = GFPGANer(
+            model_path='https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.3.pth',
+            upscale=args.outscale,
+            arch='clean',
+            channel_multiplier=2,
+            bg_upsampler=upsampler)
+    os.makedirs(args.output, exist_ok=True)
+
+    if os.path.isfile(args.input):
+        paths = [args.input]
+    else:
+        paths = sorted(glob.glob(os.path.join(args.input, '*')))
+
+    for idx, path in enumerate(paths):
+        imgname, extension = os.path.splitext(os.path.basename(path))
+        print('Testing', idx, imgname)
+
+        img = cv2.imread(path, cv2.IMREAD_COLOR)
+        if len(img.shape) == 3 and img.shape[2] == 4:
+            img_mode = 'RGBA'
+        else:
+            img_mode = None
+
+        try:
+            if args.face_enhance:
+                _, _, output = face_enhancer.enhance(img, has_aligned=False, only_center_face=False, paste_back=True)
+            else:
+                output, _ = upsampler.enhance(img, outscale=args.outscale)
+        except RuntimeError as error:
+            print('Error', error)
+            print('If you encounter CUDA out of memory, try to set --tile with a smaller number.')
+        else:
+            if args.ext == 'auto':
+                extension = extension[1:]
+            else:
+                extension = args.ext
+            if img_mode == 'RGBA':  # RGBA images should be saved in png format
+                extension = 'png'
+            if args.suffix == '':
+                save_path = os.path.join(args.output, f'{imgname}.{extension}')
+            else:
+                save_path = os.path.join(args.output, f'{imgname}_{args.suffix}.{extension}')
+            cv2.imwrite(save_path, output)
+
+
+if __name__ == '__main__':
+    main()

Real-ESRGAN/options/finetune_realesrgan_x4plus.yml

@@ -0,0 +1,188 @@
+# general settings
+name: finetune_RealESRGANx4plus_400k
+model_type: RealESRGANModel
+scale: 4
+num_gpu: auto
+manual_seed: 0
+
+# ----------------- options for synthesizing training data in RealESRGANModel ----------------- #
+# USM the ground-truth
+l1_gt_usm: True
+percep_gt_usm: True
+gan_gt_usm: False
+
+# the first degradation process
+resize_prob: [0.2, 0.7, 0.1]  # up, down, keep
+resize_range: [0.15, 1.5]
+gaussian_noise_prob: 0.5
+noise_range: [1, 30]
+poisson_scale_range: [0.05, 3]
+gray_noise_prob: 0.4
+jpeg_range: [30, 95]
+
+# the second degradation process
+second_blur_prob: 0.8
+resize_prob2: [0.3, 0.4, 0.3]  # up, down, keep
+resize_range2: [0.3, 1.2]
+gaussian_noise_prob2: 0.5
+noise_range2: [1, 25]
+poisson_scale_range2: [0.05, 2.5]
+gray_noise_prob2: 0.4
+jpeg_range2: [30, 95]
+
+gt_size: 256
+queue_size: 180
+
+# dataset and data loader settings
+datasets:
+  train:
+    name: DF2K+OST
+    type: RealESRGANDataset
+    dataroot_gt: datasets/DF2K
+    meta_info: datasets/DF2K/meta_info/meta_info_DF2Kmultiscale+OST_sub.txt
+    io_backend:
+      type: disk
+
+    blur_kernel_size: 21
+    kernel_list: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob: 0.1
+    blur_sigma: [0.2, 3]
+    betag_range: [0.5, 4]
+    betap_range: [1, 2]
+
+    blur_kernel_size2: 21
+    kernel_list2: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob2: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob2: 0.1
+    blur_sigma2: [0.2, 1.5]
+    betag_range2: [0.5, 4]
+    betap_range2: [1, 2]
+
+    final_sinc_prob: 0.8
+
+    gt_size: 256
+    use_hflip: True
+    use_rot: False
+
+    # data loader
+    use_shuffle: true
+    num_worker_per_gpu: 5
+    batch_size_per_gpu: 12
+    dataset_enlarge_ratio: 1
+    prefetch_mode: ~
+
+  # Uncomment these for validation
+  # val:
+  #   name: validation
+  #   type: PairedImageDataset
+  #   dataroot_gt: path_to_gt
+  #   dataroot_lq: path_to_lq
+  #   io_backend:
+  #     type: disk
+
+# network structures
+network_g:
+  type: RRDBNet
+  num_in_ch: 3
+  num_out_ch: 3
+  num_feat: 64
+  num_block: 23
+  num_grow_ch: 32
+
+network_d:
+  type: UNetDiscriminatorSN
+  num_in_ch: 3
+  num_feat: 64
+  skip_connection: True
+
+# path
+path:
+  # use the pre-trained Real-ESRNet model
+  pretrain_network_g: experiments/pretrained_models/RealESRNet_x4plus.pth
+  param_key_g: params_ema
+  strict_load_g: true
+  pretrain_network_d: experiments/pretrained_models/RealESRGAN_x4plus_netD.pth
+  param_key_d: params
+  strict_load_d: true
+  resume_state: ~
+
+# training settings
+train:
+  ema_decay: 0.999
+  optim_g:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+  optim_d:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+
+  scheduler:
+    type: MultiStepLR
+    milestones: [400000]
+    gamma: 0.5
+
+  total_iter: 400000
+  warmup_iter: -1  # no warm up
+
+  # losses
+  pixel_opt:
+    type: L1Loss
+    loss_weight: 1.0
+    reduction: mean
+  # perceptual loss (content and style losses)
+  perceptual_opt:
+    type: PerceptualLoss
+    layer_weights:
+      # before relu
+      'conv1_2': 0.1
+      'conv2_2': 0.1
+      'conv3_4': 1
+      'conv4_4': 1
+      'conv5_4': 1
+    vgg_type: vgg19
+    use_input_norm: true
+    perceptual_weight: !!float 1.0
+    style_weight: 0
+    range_norm: false
+    criterion: l1
+  # gan loss
+  gan_opt:
+    type: GANLoss
+    gan_type: vanilla
+    real_label_val: 1.0
+    fake_label_val: 0.0
+    loss_weight: !!float 1e-1
+
+  net_d_iters: 1
+  net_d_init_iters: 0
+
+# Uncomment these for validation
+# validation settings
+# val:
+#   val_freq: !!float 5e3
+#   save_img: True
+
+#   metrics:
+#     psnr: # metric name
+#       type: calculate_psnr
+#       crop_border: 4
+#       test_y_channel: false
+
+# logging settings
+logger:
+  print_freq: 100
+  save_checkpoint_freq: !!float 5e3
+  use_tb_logger: true
+  wandb:
+    project: ~
+    resume_id: ~
+
+# dist training settings
+dist_params:
+  backend: nccl
+  port: 29500

Real-ESRGAN/options/finetune_realesrgan_x4plus_pairdata.yml

@@ -0,0 +1,150 @@
+# general settings
+name: finetune_RealESRGANx4plus_400k_pairdata
+model_type: RealESRGANModel
+scale: 4
+num_gpu: auto
+manual_seed: 0
+
+# USM the ground-truth
+l1_gt_usm: True
+percep_gt_usm: True
+gan_gt_usm: False
+
+high_order_degradation: False # do not use the high-order degradation generation process
+
+# dataset and data loader settings
+datasets:
+  train:
+    name: DIV2K
+    type: RealESRGANPairedDataset
+    dataroot_gt: datasets/DF2K
+    dataroot_lq: datasets/DF2K
+    meta_info: datasets/DF2K/meta_info/meta_info_DIV2K_sub_pair.txt
+    io_backend:
+      type: disk
+
+    gt_size: 256
+    use_hflip: True
+    use_rot: False
+
+    # data loader
+    use_shuffle: true
+    num_worker_per_gpu: 5
+    batch_size_per_gpu: 12
+    dataset_enlarge_ratio: 1
+    prefetch_mode: ~
+
+  # Uncomment these for validation
+  # val:
+  #   name: validation
+  #   type: PairedImageDataset
+  #   dataroot_gt: path_to_gt
+  #   dataroot_lq: path_to_lq
+  #   io_backend:
+  #     type: disk
+
+# network structures
+network_g:
+  type: RRDBNet
+  num_in_ch: 3
+  num_out_ch: 3
+  num_feat: 64
+  num_block: 23
+  num_grow_ch: 32
+
+network_d:
+  type: UNetDiscriminatorSN
+  num_in_ch: 3
+  num_feat: 64
+  skip_connection: True
+
+# path
+path:
+  # use the pre-trained Real-ESRNet model
+  pretrain_network_g: experiments/pretrained_models/RealESRNet_x4plus.pth
+  param_key_g: params_ema
+  strict_load_g: true
+  pretrain_network_d: experiments/pretrained_models/RealESRGAN_x4plus_netD.pth
+  param_key_d: params
+  strict_load_d: true
+  resume_state: ~
+
+# training settings
+train:
+  ema_decay: 0.999
+  optim_g:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+  optim_d:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+
+  scheduler:
+    type: MultiStepLR
+    milestones: [400000]
+    gamma: 0.5
+
+  total_iter: 400000
+  warmup_iter: -1  # no warm up
+
+  # losses
+  pixel_opt:
+    type: L1Loss
+    loss_weight: 1.0
+    reduction: mean
+  # perceptual loss (content and style losses)
+  perceptual_opt:
+    type: PerceptualLoss
+    layer_weights:
+      # before relu
+      'conv1_2': 0.1
+      'conv2_2': 0.1
+      'conv3_4': 1
+      'conv4_4': 1
+      'conv5_4': 1
+    vgg_type: vgg19
+    use_input_norm: true
+    perceptual_weight: !!float 1.0
+    style_weight: 0
+    range_norm: false
+    criterion: l1
+  # gan loss
+  gan_opt:
+    type: GANLoss
+    gan_type: vanilla
+    real_label_val: 1.0
+    fake_label_val: 0.0
+    loss_weight: !!float 1e-1
+
+  net_d_iters: 1
+  net_d_init_iters: 0
+
+# Uncomment these for validation
+# validation settings
+# val:
+#   val_freq: !!float 5e3
+#   save_img: True
+
+#   metrics:
+#     psnr: # metric name
+#       type: calculate_psnr
+#       crop_border: 4
+#       test_y_channel: false
+
+# logging settings
+logger:
+  print_freq: 100
+  save_checkpoint_freq: !!float 5e3
+  use_tb_logger: true
+  wandb:
+    project: ~
+    resume_id: ~
+
+# dist training settings
+dist_params:
+  backend: nccl
+  port: 29500

Real-ESRGAN/options/train_realesrgan_x2plus.yml

@@ -0,0 +1,186 @@
+# general settings
+name: train_RealESRGANx2plus_400k_B12G4
+model_type: RealESRGANModel
+scale: 2
+num_gpu: auto  # auto: can infer from your visible devices automatically. official: 4 GPUs
+manual_seed: 0
+
+# ----------------- options for synthesizing training data in RealESRGANModel ----------------- #
+# USM the ground-truth
+l1_gt_usm: True
+percep_gt_usm: True
+gan_gt_usm: False
+
+# the first degradation process
+resize_prob: [0.2, 0.7, 0.1]  # up, down, keep
+resize_range: [0.15, 1.5]
+gaussian_noise_prob: 0.5
+noise_range: [1, 30]
+poisson_scale_range: [0.05, 3]
+gray_noise_prob: 0.4
+jpeg_range: [30, 95]
+
+# the second degradation process
+second_blur_prob: 0.8
+resize_prob2: [0.3, 0.4, 0.3]  # up, down, keep
+resize_range2: [0.3, 1.2]
+gaussian_noise_prob2: 0.5
+noise_range2: [1, 25]
+poisson_scale_range2: [0.05, 2.5]
+gray_noise_prob2: 0.4
+jpeg_range2: [30, 95]
+
+gt_size: 256
+queue_size: 180
+
+# dataset and data loader settings
+datasets:
+  train:
+    name: DF2K+OST
+    type: RealESRGANDataset
+    dataroot_gt: datasets/DF2K
+    meta_info: datasets/DF2K/meta_info/meta_info_DF2Kmultiscale+OST_sub.txt
+    io_backend:
+      type: disk
+
+    blur_kernel_size: 21
+    kernel_list: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob: 0.1
+    blur_sigma: [0.2, 3]
+    betag_range: [0.5, 4]
+    betap_range: [1, 2]
+
+    blur_kernel_size2: 21
+    kernel_list2: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob2: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob2: 0.1
+    blur_sigma2: [0.2, 1.5]
+    betag_range2: [0.5, 4]
+    betap_range2: [1, 2]
+
+    final_sinc_prob: 0.8
+
+    gt_size: 256
+    use_hflip: True
+    use_rot: False
+
+    # data loader
+    use_shuffle: true
+    num_worker_per_gpu: 5
+    batch_size_per_gpu: 12
+    dataset_enlarge_ratio: 1
+    prefetch_mode: ~
+
+  # Uncomment these for validation
+  # val:
+  #   name: validation
+  #   type: PairedImageDataset
+  #   dataroot_gt: path_to_gt
+  #   dataroot_lq: path_to_lq
+  #   io_backend:
+  #     type: disk
+
+# network structures
+network_g:
+  type: RRDBNet
+  num_in_ch: 3
+  num_out_ch: 3
+  num_feat: 64
+  num_block: 23
+  num_grow_ch: 32
+  scale: 2
+
+network_d:
+  type: UNetDiscriminatorSN
+  num_in_ch: 3
+  num_feat: 64
+  skip_connection: True
+
+# path
+path:
+  # use the pre-trained Real-ESRNet model
+  pretrain_network_g: experiments/pretrained_models/RealESRNet_x2plus.pth
+  param_key_g: params_ema
+  strict_load_g: true
+  resume_state: ~
+
+# training settings
+train:
+  ema_decay: 0.999
+  optim_g:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+  optim_d:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+
+  scheduler:
+    type: MultiStepLR
+    milestones: [400000]
+    gamma: 0.5
+
+  total_iter: 400000
+  warmup_iter: -1  # no warm up
+
+  # losses
+  pixel_opt:
+    type: L1Loss
+    loss_weight: 1.0
+    reduction: mean
+  # perceptual loss (content and style losses)
+  perceptual_opt:
+    type: PerceptualLoss
+    layer_weights:
+      # before relu
+      'conv1_2': 0.1
+      'conv2_2': 0.1
+      'conv3_4': 1
+      'conv4_4': 1
+      'conv5_4': 1
+    vgg_type: vgg19
+    use_input_norm: true
+    perceptual_weight: !!float 1.0
+    style_weight: 0
+    range_norm: false
+    criterion: l1
+  # gan loss
+  gan_opt:
+    type: GANLoss
+    gan_type: vanilla
+    real_label_val: 1.0
+    fake_label_val: 0.0
+    loss_weight: !!float 1e-1
+
+  net_d_iters: 1
+  net_d_init_iters: 0
+
+# Uncomment these for validation
+# validation settings
+# val:
+#   val_freq: !!float 5e3
+#   save_img: True
+
+#   metrics:
+#     psnr: # metric name
+#       type: calculate_psnr
+#       crop_border: 4
+#       test_y_channel: false
+
+# logging settings
+logger:
+  print_freq: 100
+  save_checkpoint_freq: !!float 5e3
+  use_tb_logger: true
+  wandb:
+    project: ~
+    resume_id: ~
+
+# dist training settings
+dist_params:
+  backend: nccl
+  port: 29500

Real-ESRGAN/options/train_realesrgan_x4plus.yml

@@ -0,0 +1,185 @@
+# general settings
+name: train_RealESRGANx4plus_400k_B12G4
+model_type: RealESRGANModel
+scale: 4
+num_gpu: auto  # auto: can infer from your visible devices automatically. official: 4 GPUs
+manual_seed: 0
+
+# ----------------- options for synthesizing training data in RealESRGANModel ----------------- #
+# USM the ground-truth
+l1_gt_usm: True
+percep_gt_usm: True
+gan_gt_usm: False
+
+# the first degradation process
+resize_prob: [0.2, 0.7, 0.1]  # up, down, keep
+resize_range: [0.15, 1.5]
+gaussian_noise_prob: 0.5
+noise_range: [1, 30]
+poisson_scale_range: [0.05, 3]
+gray_noise_prob: 0.4
+jpeg_range: [30, 95]
+
+# the second degradation process
+second_blur_prob: 0.8
+resize_prob2: [0.3, 0.4, 0.3]  # up, down, keep
+resize_range2: [0.3, 1.2]
+gaussian_noise_prob2: 0.5
+noise_range2: [1, 25]
+poisson_scale_range2: [0.05, 2.5]
+gray_noise_prob2: 0.4
+jpeg_range2: [30, 95]
+
+gt_size: 256
+queue_size: 180
+
+# dataset and data loader settings
+datasets:
+  train:
+    name: DF2K+OST
+    type: RealESRGANDataset
+    dataroot_gt: datasets/DF2K
+    meta_info: datasets/DF2K/meta_info/meta_info_DF2Kmultiscale+OST_sub.txt
+    io_backend:
+      type: disk
+
+    blur_kernel_size: 21
+    kernel_list: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob: 0.1
+    blur_sigma: [0.2, 3]
+    betag_range: [0.5, 4]
+    betap_range: [1, 2]
+
+    blur_kernel_size2: 21
+    kernel_list2: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob2: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob2: 0.1
+    blur_sigma2: [0.2, 1.5]
+    betag_range2: [0.5, 4]
+    betap_range2: [1, 2]
+
+    final_sinc_prob: 0.8
+
+    gt_size: 256
+    use_hflip: True
+    use_rot: False
+
+    # data loader
+    use_shuffle: true
+    num_worker_per_gpu: 5
+    batch_size_per_gpu: 12
+    dataset_enlarge_ratio: 1
+    prefetch_mode: ~
+
+  # Uncomment these for validation
+  # val:
+  #   name: validation
+  #   type: PairedImageDataset
+  #   dataroot_gt: path_to_gt
+  #   dataroot_lq: path_to_lq
+  #   io_backend:
+  #     type: disk
+
+# network structures
+network_g:
+  type: RRDBNet
+  num_in_ch: 3
+  num_out_ch: 3
+  num_feat: 64
+  num_block: 23
+  num_grow_ch: 32
+
+network_d:
+  type: UNetDiscriminatorSN
+  num_in_ch: 3
+  num_feat: 64
+  skip_connection: True
+
+# path
+path:
+  # use the pre-trained Real-ESRNet model
+  pretrain_network_g: experiments/pretrained_models/RealESRNet_x4plus.pth
+  param_key_g: params_ema
+  strict_load_g: true
+  resume_state: ~
+
+# training settings
+train:
+  ema_decay: 0.999
+  optim_g:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+  optim_d:
+    type: Adam
+    lr: !!float 1e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+
+  scheduler:
+    type: MultiStepLR
+    milestones: [400000]
+    gamma: 0.5
+
+  total_iter: 400000
+  warmup_iter: -1  # no warm up
+
+  # losses
+  pixel_opt:
+    type: L1Loss
+    loss_weight: 1.0
+    reduction: mean
+  # perceptual loss (content and style losses)
+  perceptual_opt:
+    type: PerceptualLoss
+    layer_weights:
+      # before relu
+      'conv1_2': 0.1
+      'conv2_2': 0.1
+      'conv3_4': 1
+      'conv4_4': 1
+      'conv5_4': 1
+    vgg_type: vgg19
+    use_input_norm: true
+    perceptual_weight: !!float 1.0
+    style_weight: 0
+    range_norm: false
+    criterion: l1
+  # gan loss
+  gan_opt:
+    type: GANLoss
+    gan_type: vanilla
+    real_label_val: 1.0
+    fake_label_val: 0.0
+    loss_weight: !!float 1e-1
+
+  net_d_iters: 1
+  net_d_init_iters: 0
+
+# Uncomment these for validation
+# validation settings
+# val:
+#   val_freq: !!float 5e3
+#   save_img: True
+
+#   metrics:
+#     psnr: # metric name
+#       type: calculate_psnr
+#       crop_border: 4
+#       test_y_channel: false
+
+# logging settings
+logger:
+  print_freq: 100
+  save_checkpoint_freq: !!float 5e3
+  use_tb_logger: true
+  wandb:
+    project: ~
+    resume_id: ~
+
+# dist training settings
+dist_params:
+  backend: nccl
+  port: 29500

Real-ESRGAN/options/train_realesrnet_x2plus.yml

@@ -0,0 +1,145 @@
+# general settings
+name: train_RealESRNetx2plus_1000k_B12G4
+model_type: RealESRNetModel
+scale: 2
+num_gpu: auto  # auto: can infer from your visible devices automatically. official: 4 GPUs
+manual_seed: 0
+
+# ----------------- options for synthesizing training data in RealESRNetModel ----------------- #
+gt_usm: True  # USM the ground-truth
+
+# the first degradation process
+resize_prob: [0.2, 0.7, 0.1]  # up, down, keep
+resize_range: [0.15, 1.5]
+gaussian_noise_prob: 0.5
+noise_range: [1, 30]
+poisson_scale_range: [0.05, 3]
+gray_noise_prob: 0.4
+jpeg_range: [30, 95]
+
+# the second degradation process
+second_blur_prob: 0.8
+resize_prob2: [0.3, 0.4, 0.3]  # up, down, keep
+resize_range2: [0.3, 1.2]
+gaussian_noise_prob2: 0.5
+noise_range2: [1, 25]
+poisson_scale_range2: [0.05, 2.5]
+gray_noise_prob2: 0.4
+jpeg_range2: [30, 95]
+
+gt_size: 256
+queue_size: 180
+
+# dataset and data loader settings
+datasets:
+  train:
+    name: DF2K+OST
+    type: RealESRGANDataset
+    dataroot_gt: datasets/DF2K
+    meta_info: datasets/DF2K/meta_info/meta_info_DF2Kmultiscale+OST_sub.txt
+    io_backend:
+      type: disk
+
+    blur_kernel_size: 21
+    kernel_list: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob: 0.1
+    blur_sigma: [0.2, 3]
+    betag_range: [0.5, 4]
+    betap_range: [1, 2]
+
+    blur_kernel_size2: 21
+    kernel_list2: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob2: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob2: 0.1
+    blur_sigma2: [0.2, 1.5]
+    betag_range2: [0.5, 4]
+    betap_range2: [1, 2]
+
+    final_sinc_prob: 0.8
+
+    gt_size: 256
+    use_hflip: True
+    use_rot: False
+
+    # data loader
+    use_shuffle: true
+    num_worker_per_gpu: 5
+    batch_size_per_gpu: 12
+    dataset_enlarge_ratio: 1
+    prefetch_mode: ~
+
+  # Uncomment these for validation
+  # val:
+  #   name: validation
+  #   type: PairedImageDataset
+  #   dataroot_gt: path_to_gt
+  #   dataroot_lq: path_to_lq
+  #   io_backend:
+  #     type: disk
+
+# network structures
+network_g:
+  type: RRDBNet
+  num_in_ch: 3
+  num_out_ch: 3
+  num_feat: 64
+  num_block: 23
+  num_grow_ch: 32
+  scale: 2
+
+# path
+path:
+  pretrain_network_g: experiments/pretrained_models/RealESRGAN_x4plus.pth
+  param_key_g: params_ema
+  strict_load_g: False
+  resume_state: ~
+
+# training settings
+train:
+  ema_decay: 0.999
+  optim_g:
+    type: Adam
+    lr: !!float 2e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+
+  scheduler:
+    type: MultiStepLR
+    milestones: [1000000]
+    gamma: 0.5
+
+  total_iter: 1000000
+  warmup_iter: -1  # no warm up
+
+  # losses
+  pixel_opt:
+    type: L1Loss
+    loss_weight: 1.0
+    reduction: mean
+
+# Uncomment these for validation
+# validation settings
+# val:
+#   val_freq: !!float 5e3
+#   save_img: True
+
+#   metrics:
+#     psnr: # metric name
+#       type: calculate_psnr
+#       crop_border: 4
+#       test_y_channel: false
+
+# logging settings
+logger:
+  print_freq: 100
+  save_checkpoint_freq: !!float 5e3
+  use_tb_logger: true
+  wandb:
+    project: ~
+    resume_id: ~
+
+# dist training settings
+dist_params:
+  backend: nccl
+  port: 29500

Real-ESRGAN/options/train_realesrnet_x4plus.yml

@@ -0,0 +1,144 @@
+# general settings
+name: train_RealESRNetx4plus_1000k_B12G4
+model_type: RealESRNetModel
+scale: 4
+num_gpu: auto  # auto: can infer from your visible devices automatically. official: 4 GPUs
+manual_seed: 0
+
+# ----------------- options for synthesizing training data in RealESRNetModel ----------------- #
+gt_usm: True  # USM the ground-truth
+
+# the first degradation process
+resize_prob: [0.2, 0.7, 0.1]  # up, down, keep
+resize_range: [0.15, 1.5]
+gaussian_noise_prob: 0.5
+noise_range: [1, 30]
+poisson_scale_range: [0.05, 3]
+gray_noise_prob: 0.4
+jpeg_range: [30, 95]
+
+# the second degradation process
+second_blur_prob: 0.8
+resize_prob2: [0.3, 0.4, 0.3]  # up, down, keep
+resize_range2: [0.3, 1.2]
+gaussian_noise_prob2: 0.5
+noise_range2: [1, 25]
+poisson_scale_range2: [0.05, 2.5]
+gray_noise_prob2: 0.4
+jpeg_range2: [30, 95]
+
+gt_size: 256
+queue_size: 180
+
+# dataset and data loader settings
+datasets:
+  train:
+    name: DF2K+OST
+    type: RealESRGANDataset
+    dataroot_gt: datasets/DF2K
+    meta_info: datasets/DF2K/meta_info/meta_info_DF2Kmultiscale+OST_sub.txt
+    io_backend:
+      type: disk
+
+    blur_kernel_size: 21
+    kernel_list: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob: 0.1
+    blur_sigma: [0.2, 3]
+    betag_range: [0.5, 4]
+    betap_range: [1, 2]
+
+    blur_kernel_size2: 21
+    kernel_list2: ['iso', 'aniso', 'generalized_iso', 'generalized_aniso', 'plateau_iso', 'plateau_aniso']
+    kernel_prob2: [0.45, 0.25, 0.12, 0.03, 0.12, 0.03]
+    sinc_prob2: 0.1
+    blur_sigma2: [0.2, 1.5]
+    betag_range2: [0.5, 4]
+    betap_range2: [1, 2]
+
+    final_sinc_prob: 0.8
+
+    gt_size: 256
+    use_hflip: True
+    use_rot: False
+
+    # data loader
+    use_shuffle: true
+    num_worker_per_gpu: 5
+    batch_size_per_gpu: 12
+    dataset_enlarge_ratio: 1
+    prefetch_mode: ~
+
+  # Uncomment these for validation
+  # val:
+  #   name: validation
+  #   type: PairedImageDataset
+  #   dataroot_gt: path_to_gt
+  #   dataroot_lq: path_to_lq
+  #   io_backend:
+  #     type: disk
+
+# network structures
+network_g:
+  type: RRDBNet
+  num_in_ch: 3
+  num_out_ch: 3
+  num_feat: 64
+  num_block: 23
+  num_grow_ch: 32
+
+# path
+path:
+  pretrain_network_g: experiments/pretrained_models/ESRGAN_SRx4_DF2KOST_official-ff704c30.pth
+  param_key_g: params_ema
+  strict_load_g: true
+  resume_state: ~
+
+# training settings
+train:
+  ema_decay: 0.999
+  optim_g:
+    type: Adam
+    lr: !!float 2e-4
+    weight_decay: 0
+    betas: [0.9, 0.99]
+
+  scheduler:
+    type: MultiStepLR
+    milestones: [1000000]
+    gamma: 0.5
+
+  total_iter: 1000000
+  warmup_iter: -1  # no warm up
+
+  # losses
+  pixel_opt:
+    type: L1Loss
+    loss_weight: 1.0
+    reduction: mean
+
+# Uncomment these for validation
+# validation settings
+# val:
+#   val_freq: !!float 5e3
+#   save_img: True
+
+#   metrics:
+#     psnr: # metric name
+#       type: calculate_psnr
+#       crop_border: 4
+#       test_y_channel: false
+
+# logging settings
+logger:
+  print_freq: 100
+  save_checkpoint_freq: !!float 5e3
+  use_tb_logger: true
+  wandb:
+    project: ~
+    resume_id: ~
+
+# dist training settings
+dist_params:
+  backend: nccl
+  port: 29500

Real-ESRGAN/realesrgan/__init__.py

@@ -0,0 +1,6 @@
+# flake8: noqa
+from .archs import *
+from .data import *
+from .models import *
+from .utils import *
+from .version import *

Real-ESRGAN/realesrgan/archs/__init__.py

@@ -0,0 +1,10 @@
+import importlib
+from basicsr.utils import scandir
+from os import path as osp
+
+# automatically scan and import arch modules for registry
+# scan all the files that end with '_arch.py' under the archs folder
+arch_folder = osp.dirname(osp.abspath(__file__))
+arch_filenames = [osp.splitext(osp.basename(v))[0] for v in scandir(arch_folder) if v.endswith('_arch.py')]
+# import all the arch modules
+_arch_modules = [importlib.import_module(f'realesrgan.archs.{file_name}') for file_name in arch_filenames]

Real-ESRGAN/realesrgan/archs/discriminator_arch.py

@@ -0,0 +1,67 @@
+from basicsr.utils.registry import ARCH_REGISTRY
+from torch import nn as nn
+from torch.nn import functional as F
+from torch.nn.utils import spectral_norm
+
+
+@ARCH_REGISTRY.register()
+class UNetDiscriminatorSN(nn.Module):
+    """Defines a U-Net discriminator with spectral normalization (SN)
+
+    It is used in Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data.
+
+    Arg:
+        num_in_ch (int): Channel number of inputs. Default: 3.
+        num_feat (int): Channel number of base intermediate features. Default: 64.
+        skip_connection (bool): Whether to use skip connections between U-Net. Default: True.
+    """
+
+    def __init__(self, num_in_ch, num_feat=64, skip_connection=True):
+        super(UNetDiscriminatorSN, self).__init__()
+        self.skip_connection = skip_connection
+        norm = spectral_norm
+        # the first convolution
+        self.conv0 = nn.Conv2d(num_in_ch, num_feat, kernel_size=3, stride=1, padding=1)
+        # downsample
+        self.conv1 = norm(nn.Conv2d(num_feat, num_feat * 2, 4, 2, 1, bias=False))
+        self.conv2 = norm(nn.Conv2d(num_feat * 2, num_feat * 4, 4, 2, 1, bias=False))
+        self.conv3 = norm(nn.Conv2d(num_feat * 4, num_feat * 8, 4, 2, 1, bias=False))
+        # upsample
+        self.conv4 = norm(nn.Conv2d(num_feat * 8, num_feat * 4, 3, 1, 1, bias=False))
+        self.conv5 = norm(nn.Conv2d(num_feat * 4, num_feat * 2, 3, 1, 1, bias=False))
+        self.conv6 = norm(nn.Conv2d(num_feat * 2, num_feat, 3, 1, 1, bias=False))
+        # extra convolutions
+        self.conv7 = norm(nn.Conv2d(num_feat, num_feat, 3, 1, 1, bias=False))
+        self.conv8 = norm(nn.Conv2d(num_feat, num_feat, 3, 1, 1, bias=False))
+        self.conv9 = nn.Conv2d(num_feat, 1, 3, 1, 1)
+
+    def forward(self, x):
+        # downsample
+        x0 = F.leaky_relu(self.conv0(x), negative_slope=0.2, inplace=True)
+        x1 = F.leaky_relu(self.conv1(x0), negative_slope=0.2, inplace=True)
+        x2 = F.leaky_relu(self.conv2(x1), negative_slope=0.2, inplace=True)
+        x3 = F.leaky_relu(self.conv3(x2), negative_slope=0.2, inplace=True)
+
+        # upsample
+        x3 = F.interpolate(x3, scale_factor=2, mode='bilinear', align_corners=False)
+        x4 = F.leaky_relu(self.conv4(x3), negative_slope=0.2, inplace=True)
+
+        if self.skip_connection:
+            x4 = x4 + x2
+        x4 = F.interpolate(x4, scale_factor=2, mode='bilinear', align_corners=False)
+        x5 = F.leaky_relu(self.conv5(x4), negative_slope=0.2, inplace=True)
+
+        if self.skip_connection:
+            x5 = x5 + x1
+        x5 = F.interpolate(x5, scale_factor=2, mode='bilinear', align_corners=False)
+        x6 = F.leaky_relu(self.conv6(x5), negative_slope=0.2, inplace=True)
+
+        if self.skip_connection:
+            x6 = x6 + x0
+
+        # extra convolutions
+        out = F.leaky_relu(self.conv7(x6), negative_slope=0.2, inplace=True)
+        out = F.leaky_relu(self.conv8(out), negative_slope=0.2, inplace=True)
+        out = self.conv9(out)
+
+        return out

Real-ESRGAN/realesrgan/archs/srvgg_arch.py

@@ -0,0 +1,69 @@
+from basicsr.utils.registry import ARCH_REGISTRY
+from torch import nn as nn
+from torch.nn import functional as F
+
+
+@ARCH_REGISTRY.register()
+class SRVGGNetCompact(nn.Module):
+    """A compact VGG-style network structure for super-resolution.
+
+    It is a compact network structure, which performs upsampling in the last layer and no convolution is
+    conducted on the HR feature space.
+
+    Args:
+        num_in_ch (int): Channel number of inputs. Default: 3.
+        num_out_ch (int): Channel number of outputs. Default: 3.
+        num_feat (int): Channel number of intermediate features. Default: 64.
+        num_conv (int): Number of convolution layers in the body network. Default: 16.
+        upscale (int): Upsampling factor. Default: 4.
+        act_type (str): Activation type, options: 'relu', 'prelu', 'leakyrelu'. Default: prelu.
+    """
+
+    def __init__(self, num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=16, upscale=4, act_type='prelu'):
+        super(SRVGGNetCompact, self).__init__()
+        self.num_in_ch = num_in_ch
+        self.num_out_ch = num_out_ch
+        self.num_feat = num_feat
+        self.num_conv = num_conv
+        self.upscale = upscale
+        self.act_type = act_type
+
+        self.body = nn.ModuleList()
+        # the first conv
+        self.body.append(nn.Conv2d(num_in_ch, num_feat, 3, 1, 1))
+        # the first activation
+        if act_type == 'relu':
+            activation = nn.ReLU(inplace=True)
+        elif act_type == 'prelu':
+            activation = nn.PReLU(num_parameters=num_feat)
+        elif act_type == 'leakyrelu':
+            activation = nn.LeakyReLU(negative_slope=0.1, inplace=True)
+        self.body.append(activation)
+
+        # the body structure
+        for _ in range(num_conv):
+            self.body.append(nn.Conv2d(num_feat, num_feat, 3, 1, 1))
+            # activation
+            if act_type == 'relu':
+                activation = nn.ReLU(inplace=True)
+            elif act_type == 'prelu':
+                activation = nn.PReLU(num_parameters=num_feat)
+            elif act_type == 'leakyrelu':
+                activation = nn.LeakyReLU(negative_slope=0.1, inplace=True)
+            self.body.append(activation)
+
+        # the last conv
+        self.body.append(nn.Conv2d(num_feat, num_out_ch * upscale * upscale, 3, 1, 1))
+        # upsample
+        self.upsampler = nn.PixelShuffle(upscale)
+
+    def forward(self, x):
+        out = x
+        for i in range(0, len(self.body)):
+            out = self.body[i](out)
+
+        out = self.upsampler(out)
+        # add the nearest upsampled image, so that the network learns the residual
+        base = F.interpolate(x, scale_factor=self.upscale, mode='nearest')
+        out += base
+        return out

Real-ESRGAN/realesrgan/data/__init__.py

@@ -0,0 +1,10 @@
+import importlib
+from basicsr.utils import scandir
+from os import path as osp
+
+# automatically scan and import dataset modules for registry
+# scan all the files that end with '_dataset.py' under the data folder
+data_folder = osp.dirname(osp.abspath(__file__))
+dataset_filenames = [osp.splitext(osp.basename(v))[0] for v in scandir(data_folder) if v.endswith('_dataset.py')]
+# import all the dataset modules
+_dataset_modules = [importlib.import_module(f'realesrgan.data.{file_name}') for file_name in dataset_filenames]

Real-ESRGAN/realesrgan/data/realesrgan_dataset.py

@@ -0,0 +1,192 @@
+import cv2
+import math
+import numpy as np
+import os
+import os.path as osp
+import random
+import time
+import torch
+from basicsr.data.degradations import circular_lowpass_kernel, random_mixed_kernels
+from basicsr.data.transforms import augment
+from basicsr.utils import FileClient, get_root_logger, imfrombytes, img2tensor
+from basicsr.utils.registry import DATASET_REGISTRY
+from torch.utils import data as data
+
+
+@DATASET_REGISTRY.register()
+class RealESRGANDataset(data.Dataset):
+    """Dataset used for Real-ESRGAN model:
+    Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data.
+
+    It loads gt (Ground-Truth) images, and augments them.
+    It also generates blur kernels and sinc kernels for generating low-quality images.
+    Note that the low-quality images are processed in tensors on GPUS for faster processing.
+
+    Args:
+        opt (dict): Config for train datasets. It contains the following keys:
+            dataroot_gt (str): Data root path for gt.
+            meta_info (str): Path for meta information file.
+            io_backend (dict): IO backend type and other kwarg.
+            use_hflip (bool): Use horizontal flips.
+            use_rot (bool): Use rotation (use vertical flip and transposing h and w for implementation).
+            Please see more options in the codes.
+    """
+
+    def __init__(self, opt):
+        super(RealESRGANDataset, self).__init__()
+        self.opt = opt
+        self.file_client = None
+        self.io_backend_opt = opt['io_backend']
+        self.gt_folder = opt['dataroot_gt']
+
+        # file client (lmdb io backend)
+        if self.io_backend_opt['type'] == 'lmdb':
+            self.io_backend_opt['db_paths'] = [self.gt_folder]
+            self.io_backend_opt['client_keys'] = ['gt']
+            if not self.gt_folder.endswith('.lmdb'):
+                raise ValueError(f"'dataroot_gt' should end with '.lmdb', but received {self.gt_folder}")
+            with open(osp.join(self.gt_folder, 'meta_info.txt')) as fin:
+                self.paths = [line.split('.')[0] for line in fin]
+        else:
+            # disk backend with meta_info
+            # Each line in the meta_info describes the relative path to an image
+            with open(self.opt['meta_info']) as fin:
+                paths = [line.strip().split(' ')[0] for line in fin]
+                self.paths = [os.path.join(self.gt_folder, v) for v in paths]
+
+        # blur settings for the first degradation
+        self.blur_kernel_size = opt['blur_kernel_size']
+        self.kernel_list = opt['kernel_list']
+        self.kernel_prob = opt['kernel_prob']  # a list for each kernel probability
+        self.blur_sigma = opt['blur_sigma']
+        self.betag_range = opt['betag_range']  # betag used in generalized Gaussian blur kernels
+        self.betap_range = opt['betap_range']  # betap used in plateau blur kernels
+        self.sinc_prob = opt['sinc_prob']  # the probability for sinc filters
+
+        # blur settings for the second degradation
+        self.blur_kernel_size2 = opt['blur_kernel_size2']
+        self.kernel_list2 = opt['kernel_list2']
+        self.kernel_prob2 = opt['kernel_prob2']
+        self.blur_sigma2 = opt['blur_sigma2']
+        self.betag_range2 = opt['betag_range2']
+        self.betap_range2 = opt['betap_range2']
+        self.sinc_prob2 = opt['sinc_prob2']
+
+        # a final sinc filter
+        self.final_sinc_prob = opt['final_sinc_prob']
+
+        self.kernel_range = [2 * v + 1 for v in range(3, 11)]  # kernel size ranges from 7 to 21
+        # TODO: kernel range is now hard-coded, should be in the configure file
+        self.pulse_tensor = torch.zeros(21, 21).float()  # convolving with pulse tensor brings no blurry effect
+        self.pulse_tensor[10, 10] = 1
+
+    def __getitem__(self, index):
+        if self.file_client is None:
+            self.file_client = FileClient(self.io_backend_opt.pop('type'), **self.io_backend_opt)
+
+        # -------------------------------- Load gt images -------------------------------- #
+        # Shape: (h, w, c); channel order: BGR; image range: [0, 1], float32.
+        gt_path = self.paths[index]
+        # avoid errors caused by high latency in reading files
+        retry = 3
+        while retry > 0:
+            try:
+                img_bytes = self.file_client.get(gt_path, 'gt')
+            except (IOError, OSError) as e:
+                logger = get_root_logger()
+                logger.warn(f'File client error: {e}, remaining retry times: {retry - 1}')
+                # change another file to read
+                index = random.randint(0, self.__len__())
+                gt_path = self.paths[index]
+                time.sleep(1)  # sleep 1s for occasional server congestion
+            else:
+                break
+            finally:
+                retry -= 1
+        img_gt = imfrombytes(img_bytes, float32=True)
+
+        # -------------------- Do augmentation for training: flip, rotation -------------------- #
+        img_gt = augment(img_gt, self.opt['use_hflip'], self.opt['use_rot'])
+
+        # crop or pad to 400
+        # TODO: 400 is hard-coded. You may change it accordingly
+        h, w = img_gt.shape[0:2]
+        crop_pad_size = 400
+        # pad
+        if h < crop_pad_size or w < crop_pad_size:
+            pad_h = max(0, crop_pad_size - h)
+            pad_w = max(0, crop_pad_size - w)
+            img_gt = cv2.copyMakeBorder(img_gt, 0, pad_h, 0, pad_w, cv2.BORDER_REFLECT_101)
+        # crop
+        if img_gt.shape[0] > crop_pad_size or img_gt.shape[1] > crop_pad_size:
+            h, w = img_gt.shape[0:2]
+            # randomly choose top and left coordinates
+            top = random.randint(0, h - crop_pad_size)
+            left = random.randint(0, w - crop_pad_size)
+            img_gt = img_gt[top:top + crop_pad_size, left:left + crop_pad_size, ...]
+
+        # ------------------------ Generate kernels (used in the first degradation) ------------------------ #
+        kernel_size = random.choice(self.kernel_range)
+        if np.random.uniform() < self.opt['sinc_prob']:
+            # this sinc filter setting is for kernels ranging from [7, 21]
+            if kernel_size < 13:
+                omega_c = np.random.uniform(np.pi / 3, np.pi)
+            else:
+                omega_c = np.random.uniform(np.pi / 5, np.pi)
+            kernel = circular_lowpass_kernel(omega_c, kernel_size, pad_to=False)
+        else:
+            kernel = random_mixed_kernels(
+                self.kernel_list,
+                self.kernel_prob,
+                kernel_size,
+                self.blur_sigma,
+                self.blur_sigma, [-math.pi, math.pi],
+                self.betag_range,
+                self.betap_range,
+                noise_range=None)
+        # pad kernel
+        pad_size = (21 - kernel_size) // 2
+        kernel = np.pad(kernel, ((pad_size, pad_size), (pad_size, pad_size)))
+
+        # ------------------------ Generate kernels (used in the second degradation) ------------------------ #
+        kernel_size = random.choice(self.kernel_range)
+        if np.random.uniform() < self.opt['sinc_prob2']:
+            if kernel_size < 13:
+                omega_c = np.random.uniform(np.pi / 3, np.pi)
+            else:
+                omega_c = np.random.uniform(np.pi / 5, np.pi)
+            kernel2 = circular_lowpass_kernel(omega_c, kernel_size, pad_to=False)
+        else:
+            kernel2 = random_mixed_kernels(
+                self.kernel_list2,
+                self.kernel_prob2,
+                kernel_size,
+                self.blur_sigma2,
+                self.blur_sigma2, [-math.pi, math.pi],
+                self.betag_range2,
+                self.betap_range2,
+                noise_range=None)
+
+        # pad kernel
+        pad_size = (21 - kernel_size) // 2
+        kernel2 = np.pad(kernel2, ((pad_size, pad_size), (pad_size, pad_size)))
+
+        # ------------------------------------- the final sinc kernel ------------------------------------- #
+        if np.random.uniform() < self.opt['final_sinc_prob']:
+            kernel_size = random.choice(self.kernel_range)
+            omega_c = np.random.uniform(np.pi / 3, np.pi)
+            sinc_kernel = circular_lowpass_kernel(omega_c, kernel_size, pad_to=21)
+            sinc_kernel = torch.FloatTensor(sinc_kernel)
+        else:
+            sinc_kernel = self.pulse_tensor
+
+        # BGR to RGB, HWC to CHW, numpy to tensor
+        img_gt = img2tensor([img_gt], bgr2rgb=True, float32=True)[0]
+        kernel = torch.FloatTensor(kernel)
+        kernel2 = torch.FloatTensor(kernel2)
+
+        return_d = {'gt': img_gt, 'kernel1': kernel, 'kernel2': kernel2, 'sinc_kernel': sinc_kernel, 'gt_path': gt_path}
+        return return_d
+
+    def __len__(self):
+        return len(self.paths)

Real-ESRGAN/realesrgan/data/realesrgan_paired_dataset.py

@@ -0,0 +1,108 @@
+import os
+from basicsr.data.data_util import paired_paths_from_folder, paired_paths_from_lmdb
+from basicsr.data.transforms import augment, paired_random_crop
+from basicsr.utils import FileClient, imfrombytes, img2tensor
+from basicsr.utils.registry import DATASET_REGISTRY
+from torch.utils import data as data
+from torchvision.transforms.functional import normalize
+
+
+@DATASET_REGISTRY.register()
+class RealESRGANPairedDataset(data.Dataset):
+    """Paired image dataset for image restoration.
+
+    Read LQ (Low Quality, e.g. LR (Low Resolution), blurry, noisy, etc) and GT image pairs.
+
+    There are three modes:
+    1. 'lmdb': Use lmdb files.
+        If opt['io_backend'] == lmdb.
+    2. 'meta_info': Use meta information file to generate paths.
+        If opt['io_backend'] != lmdb and opt['meta_info'] is not None.
+    3. 'folder': Scan folders to generate paths.
+        The rest.
+
+    Args:
+        opt (dict): Config for train datasets. It contains the following keys:
+            dataroot_gt (str): Data root path for gt.
+            dataroot_lq (str): Data root path for lq.
+            meta_info (str): Path for meta information file.
+            io_backend (dict): IO backend type and other kwarg.
+            filename_tmpl (str): Template for each filename. Note that the template excludes the file extension.
+                Default: '{}'.
+            gt_size (int): Cropped patched size for gt patches.
+            use_hflip (bool): Use horizontal flips.
+            use_rot (bool): Use rotation (use vertical flip and transposing h
+                and w for implementation).
+
+            scale (bool): Scale, which will be added automatically.
+            phase (str): 'train' or 'val'.
+    """
+
+    def __init__(self, opt):
+        super(RealESRGANPairedDataset, self).__init__()
+        self.opt = opt
+        self.file_client = None
+        self.io_backend_opt = opt['io_backend']
+        # mean and std for normalizing the input images
+        self.mean = opt['mean'] if 'mean' in opt else None
+        self.std = opt['std'] if 'std' in opt else None
+
+        self.gt_folder, self.lq_folder = opt['dataroot_gt'], opt['dataroot_lq']
+        self.filename_tmpl = opt['filename_tmpl'] if 'filename_tmpl' in opt else '{}'
+
+        # file client (lmdb io backend)
+        if self.io_backend_opt['type'] == 'lmdb':
+            self.io_backend_opt['db_paths'] = [self.lq_folder, self.gt_folder]
+            self.io_backend_opt['client_keys'] = ['lq', 'gt']
+            self.paths = paired_paths_from_lmdb([self.lq_folder, self.gt_folder], ['lq', 'gt'])
+        elif 'meta_info' in self.opt and self.opt['meta_info'] is not None:
+            # disk backend with meta_info
+            # Each line in the meta_info describes the relative path to an image
+            with open(self.opt['meta_info']) as fin:
+                paths = [line.strip() for line in fin]
+            self.paths = []
+            for path in paths:
+                gt_path, lq_path = path.split(', ')
+                gt_path = os.path.join(self.gt_folder, gt_path)
+                lq_path = os.path.join(self.lq_folder, lq_path)
+                self.paths.append(dict([('gt_path', gt_path), ('lq_path', lq_path)]))
+        else:
+            # disk backend
+            # it will scan the whole folder to get meta info
+            # it will be time-consuming for folders with too many files. It is recommended using an extra meta txt file
+            self.paths = paired_paths_from_folder([self.lq_folder, self.gt_folder], ['lq', 'gt'], self.filename_tmpl)
+
+    def __getitem__(self, index):
+        if self.file_client is None:
+            self.file_client = FileClient(self.io_backend_opt.pop('type'), **self.io_backend_opt)
+
+        scale = self.opt['scale']
+
+        # Load gt and lq images. Dimension order: HWC; channel order: BGR;
+        # image range: [0, 1], float32.
+        gt_path = self.paths[index]['gt_path']
+        img_bytes = self.file_client.get(gt_path, 'gt')
+        img_gt = imfrombytes(img_bytes, float32=True)
+        lq_path = self.paths[index]['lq_path']
+        img_bytes = self.file_client.get(lq_path, 'lq')
+        img_lq = imfrombytes(img_bytes, float32=True)
+
+        # augmentation for training
+        if self.opt['phase'] == 'train':
+            gt_size = self.opt['gt_size']
+            # random crop
+            img_gt, img_lq = paired_random_crop(img_gt, img_lq, gt_size, scale, gt_path)
+            # flip, rotation
+            img_gt, img_lq = augment([img_gt, img_lq], self.opt['use_hflip'], self.opt['use_rot'])
+
+        # BGR to RGB, HWC to CHW, numpy to tensor
+        img_gt, img_lq = img2tensor([img_gt, img_lq], bgr2rgb=True, float32=True)
+        # normalize
+        if self.mean is not None or self.std is not None:
+            normalize(img_lq, self.mean, self.std, inplace=True)
+            normalize(img_gt, self.mean, self.std, inplace=True)
+
+        return {'lq': img_lq, 'gt': img_gt, 'lq_path': lq_path, 'gt_path': gt_path}
+
+    def __len__(self):
+        return len(self.paths)

Real-ESRGAN/realesrgan/models/__init__.py

@@ -0,0 +1,10 @@
+import importlib
+from basicsr.utils import scandir
+from os import path as osp
+
+# automatically scan and import model modules for registry
+# scan all the files that end with '_model.py' under the model folder
+model_folder = osp.dirname(osp.abspath(__file__))
+model_filenames = [osp.splitext(osp.basename(v))[0] for v in scandir(model_folder) if v.endswith('_model.py')]
+# import all the model modules
+_model_modules = [importlib.import_module(f'realesrgan.models.{file_name}') for file_name in model_filenames]

Real-ESRGAN/realesrgan/models/realesrgan_model.py

@@ -0,0 +1,258 @@
+import numpy as np
+import random
+import torch
+from basicsr.data.degradations import random_add_gaussian_noise_pt, random_add_poisson_noise_pt
+from basicsr.data.transforms import paired_random_crop
+from basicsr.models.srgan_model import SRGANModel
+from basicsr.utils import DiffJPEG, USMSharp
+from basicsr.utils.img_process_util import filter2D
+from basicsr.utils.registry import MODEL_REGISTRY
+from collections import OrderedDict
+from torch.nn import functional as F
+
+
+@MODEL_REGISTRY.register()
+class RealESRGANModel(SRGANModel):
+    """RealESRGAN Model for Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data.
+
+    It mainly performs:
+    1. randomly synthesize LQ images in GPU tensors
+    2. optimize the networks with GAN training.
+    """
+
+    def __init__(self, opt):
+        super(RealESRGANModel, self).__init__(opt)
+        self.jpeger = DiffJPEG(differentiable=False).cuda()  # simulate JPEG compression artifacts
+        self.usm_sharpener = USMSharp().cuda()  # do usm sharpening
+        self.queue_size = opt.get('queue_size', 180)
+
+    @torch.no_grad()
+    def _dequeue_and_enqueue(self):
+        """It is the training pair pool for increasing the diversity in a batch.
+
+        Batch processing limits the diversity of synthetic degradations in a batch. For example, samples in a
+        batch could not have different resize scaling factors. Therefore, we employ this training pair pool
+        to increase the degradation diversity in a batch.
+        """
+        # initialize
+        b, c, h, w = self.lq.size()
+        if not hasattr(self, 'queue_lr'):
+            assert self.queue_size % b == 0, f'queue size {self.queue_size} should be divisible by batch size {b}'
+            self.queue_lr = torch.zeros(self.queue_size, c, h, w).cuda()
+            _, c, h, w = self.gt.size()
+            self.queue_gt = torch.zeros(self.queue_size, c, h, w).cuda()
+            self.queue_ptr = 0
+        if self.queue_ptr == self.queue_size:  # the pool is full
+            # do dequeue and enqueue
+            # shuffle
+            idx = torch.randperm(self.queue_size)
+            self.queue_lr = self.queue_lr[idx]
+            self.queue_gt = self.queue_gt[idx]
+            # get first b samples
+            lq_dequeue = self.queue_lr[0:b, :, :, :].clone()
+            gt_dequeue = self.queue_gt[0:b, :, :, :].clone()
+            # update the queue
+            self.queue_lr[0:b, :, :, :] = self.lq.clone()
+            self.queue_gt[0:b, :, :, :] = self.gt.clone()
+
+            self.lq = lq_dequeue
+            self.gt = gt_dequeue
+        else:
+            # only do enqueue
+            self.queue_lr[self.queue_ptr:self.queue_ptr + b, :, :, :] = self.lq.clone()
+            self.queue_gt[self.queue_ptr:self.queue_ptr + b, :, :, :] = self.gt.clone()
+            self.queue_ptr = self.queue_ptr + b
+
+    @torch.no_grad()
+    def feed_data(self, data):
+        """Accept data from dataloader, and then add two-order degradations to obtain LQ images.
+        """
+        if self.is_train and self.opt.get('high_order_degradation', True):
+            # training data synthesis
+            self.gt = data['gt'].to(self.device)
+            self.gt_usm = self.usm_sharpener(self.gt)
+
+            self.kernel1 = data['kernel1'].to(self.device)
+            self.kernel2 = data['kernel2'].to(self.device)
+            self.sinc_kernel = data['sinc_kernel'].to(self.device)
+
+            ori_h, ori_w = self.gt.size()[2:4]
+
+            # ----------------------- The first degradation process ----------------------- #
+            # blur
+            out = filter2D(self.gt_usm, self.kernel1)
+            # random resize
+            updown_type = random.choices(['up', 'down', 'keep'], self.opt['resize_prob'])[0]
+            if updown_type == 'up':
+                scale = np.random.uniform(1, self.opt['resize_range'][1])
+            elif updown_type == 'down':
+                scale = np.random.uniform(self.opt['resize_range'][0], 1)
+            else:
+                scale = 1
+            mode = random.choice(['area', 'bilinear', 'bicubic'])
+            out = F.interpolate(out, scale_factor=scale, mode=mode)
+            # add noise
+            gray_noise_prob = self.opt['gray_noise_prob']
+            if np.random.uniform() < self.opt['gaussian_noise_prob']:
+                out = random_add_gaussian_noise_pt(
+                    out, sigma_range=self.opt['noise_range'], clip=True, rounds=False, gray_prob=gray_noise_prob)
+            else:
+                out = random_add_poisson_noise_pt(
+                    out,
+                    scale_range=self.opt['poisson_scale_range'],
+                    gray_prob=gray_noise_prob,
+                    clip=True,
+                    rounds=False)
+            # JPEG compression
+            jpeg_p = out.new_zeros(out.size(0)).uniform_(*self.opt['jpeg_range'])
+            out = torch.clamp(out, 0, 1)  # clamp to [0, 1], otherwise JPEGer will result in unpleasant artifacts
+            out = self.jpeger(out, quality=jpeg_p)
+
+            # ----------------------- The second degradation process ----------------------- #
+            # blur
+            if np.random.uniform() < self.opt['second_blur_prob']:
+                out = filter2D(out, self.kernel2)
+            # random resize
+            updown_type = random.choices(['up', 'down', 'keep'], self.opt['resize_prob2'])[0]
+            if updown_type == 'up':
+                scale = np.random.uniform(1, self.opt['resize_range2'][1])
+            elif updown_type == 'down':
+                scale = np.random.uniform(self.opt['resize_range2'][0], 1)
+            else:
+                scale = 1
+            mode = random.choice(['area', 'bilinear', 'bicubic'])
+            out = F.interpolate(
+                out, size=(int(ori_h / self.opt['scale'] * scale), int(ori_w / self.opt['scale'] * scale)), mode=mode)
+            # add noise
+            gray_noise_prob = self.opt['gray_noise_prob2']
+            if np.random.uniform() < self.opt['gaussian_noise_prob2']:
+                out = random_add_gaussian_noise_pt(
+                    out, sigma_range=self.opt['noise_range2'], clip=True, rounds=False, gray_prob=gray_noise_prob)
+            else:
+                out = random_add_poisson_noise_pt(
+                    out,
+                    scale_range=self.opt['poisson_scale_range2'],
+                    gray_prob=gray_noise_prob,
+                    clip=True,
+                    rounds=False)
+
+            # JPEG compression + the final sinc filter
+            # We also need to resize images to desired sizes. We group [resize back + sinc filter] together
+            # as one operation.
+            # We consider two orders:
+            #   1. [resize back + sinc filter] + JPEG compression
+            #   2. JPEG compression + [resize back + sinc filter]
+            # Empirically, we find other combinations (sinc + JPEG + Resize) will introduce twisted lines.
+            if np.random.uniform() < 0.5:
+                # resize back + the final sinc filter
+                mode = random.choice(['area', 'bilinear', 'bicubic'])
+                out = F.interpolate(out, size=(ori_h // self.opt['scale'], ori_w // self.opt['scale']), mode=mode)
+                out = filter2D(out, self.sinc_kernel)
+                # JPEG compression
+                jpeg_p = out.new_zeros(out.size(0)).uniform_(*self.opt['jpeg_range2'])
+                out = torch.clamp(out, 0, 1)
+                out = self.jpeger(out, quality=jpeg_p)
+            else:
+                # JPEG compression
+                jpeg_p = out.new_zeros(out.size(0)).uniform_(*self.opt['jpeg_range2'])
+                out = torch.clamp(out, 0, 1)
+                out = self.jpeger(out, quality=jpeg_p)
+                # resize back + the final sinc filter
+                mode = random.choice(['area', 'bilinear', 'bicubic'])
+                out = F.interpolate(out, size=(ori_h // self.opt['scale'], ori_w // self.opt['scale']), mode=mode)
+                out = filter2D(out, self.sinc_kernel)
+
+            # clamp and round
+            self.lq = torch.clamp((out * 255.0).round(), 0, 255) / 255.
+
+            # random crop
+            gt_size = self.opt['gt_size']
+            (self.gt, self.gt_usm), self.lq = paired_random_crop([self.gt, self.gt_usm], self.lq, gt_size,
+                                                                 self.opt['scale'])
+
+            # training pair pool
+            self._dequeue_and_enqueue()
+            # sharpen self.gt again, as we have changed the self.gt with self._dequeue_and_enqueue
+            self.gt_usm = self.usm_sharpener(self.gt)
+            self.lq = self.lq.contiguous()  # for the warning: grad and param do not obey the gradient layout contract
+        else:
+            # for paired training or validation
+            self.lq = data['lq'].to(self.device)
+            if 'gt' in data:
+                self.gt = data['gt'].to(self.device)
+                self.gt_usm = self.usm_sharpener(self.gt)
+
+    def nondist_validation(self, dataloader, current_iter, tb_logger, save_img):
+        # do not use the synthetic process during validation
+        self.is_train = False
+        super(RealESRGANModel, self).nondist_validation(dataloader, current_iter, tb_logger, save_img)
+        self.is_train = True
+
+    def optimize_parameters(self, current_iter):
+        # usm sharpening
+        l1_gt = self.gt_usm
+        percep_gt = self.gt_usm
+        gan_gt = self.gt_usm
+        if self.opt['l1_gt_usm'] is False:
+            l1_gt = self.gt
+        if self.opt['percep_gt_usm'] is False:
+            percep_gt = self.gt
+        if self.opt['gan_gt_usm'] is False:
+            gan_gt = self.gt
+
+        # optimize net_g
+        for p in self.net_d.parameters():
+            p.requires_grad = False
+
+        self.optimizer_g.zero_grad()
+        self.output = self.net_g(self.lq)
+
+        l_g_total = 0
+        loss_dict = OrderedDict()
+        if (current_iter % self.net_d_iters == 0 and current_iter > self.net_d_init_iters):
+            # pixel loss
+            if self.cri_pix:
+                l_g_pix = self.cri_pix(self.output, l1_gt)
+                l_g_total += l_g_pix
+                loss_dict['l_g_pix'] = l_g_pix
+            # perceptual loss
+            if self.cri_perceptual:
+                l_g_percep, l_g_style = self.cri_perceptual(self.output, percep_gt)
+                if l_g_percep is not None:
+                    l_g_total += l_g_percep
+                    loss_dict['l_g_percep'] = l_g_percep
+                if l_g_style is not None:
+                    l_g_total += l_g_style
+                    loss_dict['l_g_style'] = l_g_style
+            # gan loss
+            fake_g_pred = self.net_d(self.output)
+            l_g_gan = self.cri_gan(fake_g_pred, True, is_disc=False)
+            l_g_total += l_g_gan
+            loss_dict['l_g_gan'] = l_g_gan
+
+            l_g_total.backward()
+            self.optimizer_g.step()
+
+        # optimize net_d
+        for p in self.net_d.parameters():
+            p.requires_grad = True
+
+        self.optimizer_d.zero_grad()
+        # real
+        real_d_pred = self.net_d(gan_gt)
+        l_d_real = self.cri_gan(real_d_pred, True, is_disc=True)
+        loss_dict['l_d_real'] = l_d_real
+        loss_dict['out_d_real'] = torch.mean(real_d_pred.detach())
+        l_d_real.backward()
+        # fake
+        fake_d_pred = self.net_d(self.output.detach().clone())  # clone for pt1.9
+        l_d_fake = self.cri_gan(fake_d_pred, False, is_disc=True)
+        loss_dict['l_d_fake'] = l_d_fake
+        loss_dict['out_d_fake'] = torch.mean(fake_d_pred.detach())
+        l_d_fake.backward()
+        self.optimizer_d.step()
+
+        if self.ema_decay > 0:
+            self.model_ema(decay=self.ema_decay)
+
+        self.log_dict = self.reduce_loss_dict(loss_dict)

Real-ESRGAN/realesrgan/models/realesrnet_model.py

@@ -0,0 +1,188 @@
+import numpy as np
+import random
+import torch
+from basicsr.data.degradations import random_add_gaussian_noise_pt, random_add_poisson_noise_pt
+from basicsr.data.transforms import paired_random_crop
+from basicsr.models.sr_model import SRModel
+from basicsr.utils import DiffJPEG, USMSharp
+from basicsr.utils.img_process_util import filter2D
+from basicsr.utils.registry import MODEL_REGISTRY
+from torch.nn import functional as F
+
+
+@MODEL_REGISTRY.register()
+class RealESRNetModel(SRModel):
+    """RealESRNet Model for Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data.
+
+    It is trained without GAN losses.
+    It mainly performs:
+    1. randomly synthesize LQ images in GPU tensors
+    2. optimize the networks with GAN training.
+    """
+
+    def __init__(self, opt):
+        super(RealESRNetModel, self).__init__(opt)
+        self.jpeger = DiffJPEG(differentiable=False).cuda()  # simulate JPEG compression artifacts
+        self.usm_sharpener = USMSharp().cuda()  # do usm sharpening
+        self.queue_size = opt.get('queue_size', 180)
+
+    @torch.no_grad()
+    def _dequeue_and_enqueue(self):
+        """It is the training pair pool for increasing the diversity in a batch.
+
+        Batch processing limits the diversity of synthetic degradations in a batch. For example, samples in a
+        batch could not have different resize scaling factors. Therefore, we employ this training pair pool
+        to increase the degradation diversity in a batch.
+        """
+        # initialize
+        b, c, h, w = self.lq.size()
+        if not hasattr(self, 'queue_lr'):
+            assert self.queue_size % b == 0, f'queue size {self.queue_size} should be divisible by batch size {b}'
+            self.queue_lr = torch.zeros(self.queue_size, c, h, w).cuda()
+            _, c, h, w = self.gt.size()
+            self.queue_gt = torch.zeros(self.queue_size, c, h, w).cuda()
+            self.queue_ptr = 0
+        if self.queue_ptr == self.queue_size:  # the pool is full
+            # do dequeue and enqueue
+            # shuffle
+            idx = torch.randperm(self.queue_size)
+            self.queue_lr = self.queue_lr[idx]
+            self.queue_gt = self.queue_gt[idx]
+            # get first b samples
+            lq_dequeue = self.queue_lr[0:b, :, :, :].clone()
+            gt_dequeue = self.queue_gt[0:b, :, :, :].clone()
+            # update the queue
+            self.queue_lr[0:b, :, :, :] = self.lq.clone()
+            self.queue_gt[0:b, :, :, :] = self.gt.clone()
+
+            self.lq = lq_dequeue
+            self.gt = gt_dequeue
+        else:
+            # only do enqueue
+            self.queue_lr[self.queue_ptr:self.queue_ptr + b, :, :, :] = self.lq.clone()
+            self.queue_gt[self.queue_ptr:self.queue_ptr + b, :, :, :] = self.gt.clone()
+            self.queue_ptr = self.queue_ptr + b
+
+    @torch.no_grad()
+    def feed_data(self, data):
+        """Accept data from dataloader, and then add two-order degradations to obtain LQ images.
+        """
+        if self.is_train and self.opt.get('high_order_degradation', True):
+            # training data synthesis
+            self.gt = data['gt'].to(self.device)
+            # USM sharpen the GT images
+            if self.opt['gt_usm'] is True:
+                self.gt = self.usm_sharpener(self.gt)
+
+            self.kernel1 = data['kernel1'].to(self.device)
+            self.kernel2 = data['kernel2'].to(self.device)
+            self.sinc_kernel = data['sinc_kernel'].to(self.device)
+
+            ori_h, ori_w = self.gt.size()[2:4]
+
+            # ----------------------- The first degradation process ----------------------- #
+            # blur
+            out = filter2D(self.gt, self.kernel1)
+            # random resize
+            updown_type = random.choices(['up', 'down', 'keep'], self.opt['resize_prob'])[0]
+            if updown_type == 'up':
+                scale = np.random.uniform(1, self.opt['resize_range'][1])
+            elif updown_type == 'down':
+                scale = np.random.uniform(self.opt['resize_range'][0], 1)
+            else:
+                scale = 1
+            mode = random.choice(['area', 'bilinear', 'bicubic'])
+            out = F.interpolate(out, scale_factor=scale, mode=mode)
+            # add noise
+            gray_noise_prob = self.opt['gray_noise_prob']
+            if np.random.uniform() < self.opt['gaussian_noise_prob']:
+                out = random_add_gaussian_noise_pt(
+                    out, sigma_range=self.opt['noise_range'], clip=True, rounds=False, gray_prob=gray_noise_prob)
+            else:
+                out = random_add_poisson_noise_pt(
+                    out,
+                    scale_range=self.opt['poisson_scale_range'],
+                    gray_prob=gray_noise_prob,
+                    clip=True,
+                    rounds=False)
+            # JPEG compression
+            jpeg_p = out.new_zeros(out.size(0)).uniform_(*self.opt['jpeg_range'])
+            out = torch.clamp(out, 0, 1)  # clamp to [0, 1], otherwise JPEGer will result in unpleasant artifacts
+            out = self.jpeger(out, quality=jpeg_p)
+
+            # ----------------------- The second degradation process ----------------------- #
+            # blur
+            if np.random.uniform() < self.opt['second_blur_prob']:
+                out = filter2D(out, self.kernel2)
+            # random resize
+            updown_type = random.choices(['up', 'down', 'keep'], self.opt['resize_prob2'])[0]
+            if updown_type == 'up':
+                scale = np.random.uniform(1, self.opt['resize_range2'][1])
+            elif updown_type == 'down':
+                scale = np.random.uniform(self.opt['resize_range2'][0], 1)
+            else:
+                scale = 1
+            mode = random.choice(['area', 'bilinear', 'bicubic'])
+            out = F.interpolate(
+                out, size=(int(ori_h / self.opt['scale'] * scale), int(ori_w / self.opt['scale'] * scale)), mode=mode)
+            # add noise
+            gray_noise_prob = self.opt['gray_noise_prob2']
+            if np.random.uniform() < self.opt['gaussian_noise_prob2']:
+                out = random_add_gaussian_noise_pt(
+                    out, sigma_range=self.opt['noise_range2'], clip=True, rounds=False, gray_prob=gray_noise_prob)
+            else:
+                out = random_add_poisson_noise_pt(
+                    out,
+                    scale_range=self.opt['poisson_scale_range2'],
+                    gray_prob=gray_noise_prob,
+                    clip=True,
+                    rounds=False)
+
+            # JPEG compression + the final sinc filter
+            # We also need to resize images to desired sizes. We group [resize back + sinc filter] together
+            # as one operation.
+            # We consider two orders:
+            #   1. [resize back + sinc filter] + JPEG compression
+            #   2. JPEG compression + [resize back + sinc filter]
+            # Empirically, we find other combinations (sinc + JPEG + Resize) will introduce twisted lines.
+            if np.random.uniform() < 0.5:
+                # resize back + the final sinc filter
+                mode = random.choice(['area', 'bilinear', 'bicubic'])
+                out = F.interpolate(out, size=(ori_h // self.opt['scale'], ori_w // self.opt['scale']), mode=mode)
+                out = filter2D(out, self.sinc_kernel)
+                # JPEG compression
+                jpeg_p = out.new_zeros(out.size(0)).uniform_(*self.opt['jpeg_range2'])
+                out = torch.clamp(out, 0, 1)
+                out = self.jpeger(out, quality=jpeg_p)
+            else:
+                # JPEG compression
+                jpeg_p = out.new_zeros(out.size(0)).uniform_(*self.opt['jpeg_range2'])
+                out = torch.clamp(out, 0, 1)
+                out = self.jpeger(out, quality=jpeg_p)
+                # resize back + the final sinc filter
+                mode = random.choice(['area', 'bilinear', 'bicubic'])
+                out = F.interpolate(out, size=(ori_h // self.opt['scale'], ori_w // self.opt['scale']), mode=mode)
+                out = filter2D(out, self.sinc_kernel)
+
+            # clamp and round
+            self.lq = torch.clamp((out * 255.0).round(), 0, 255) / 255.
+
+            # random crop
+            gt_size = self.opt['gt_size']
+            self.gt, self.lq = paired_random_crop(self.gt, self.lq, gt_size, self.opt['scale'])
+
+            # training pair pool
+            self._dequeue_and_enqueue()
+            self.lq = self.lq.contiguous()  # for the warning: grad and param do not obey the gradient layout contract
+        else:
+            # for paired training or validation
+            self.lq = data['lq'].to(self.device)
+            if 'gt' in data:
+                self.gt = data['gt'].to(self.device)
+                self.gt_usm = self.usm_sharpener(self.gt)
+
+    def nondist_validation(self, dataloader, current_iter, tb_logger, save_img):
+        # do not use the synthetic process during validation
+        self.is_train = False
+        super(RealESRNetModel, self).nondist_validation(dataloader, current_iter, tb_logger, save_img)
+        self.is_train = True

Real-ESRGAN/realesrgan/train.py

@@ -0,0 +1,11 @@
+# flake8: noqa
+import os.path as osp
+from basicsr.train import train_pipeline
+
+import realesrgan.archs
+import realesrgan.data
+import realesrgan.models
+
+if __name__ == '__main__':
+    root_path = osp.abspath(osp.join(__file__, osp.pardir, osp.pardir))
+    train_pipeline(root_path)

Real-ESRGAN/realesrgan/utils.py

@@ -0,0 +1,313 @@
+import cv2
+import math
+import numpy as np
+import os
+import queue
+import threading
+import torch
+from basicsr.utils.download_util import load_file_from_url
+from torch.nn import functional as F
+
+ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+
+class RealESRGANer():
+    """A helper class for upsampling images with RealESRGAN.
+
+    Args:
+        scale (int): Upsampling scale factor used in the networks. It is usually 2 or 4.
+        model_path (str): The path to the pretrained model. It can be urls (will first download it automatically).
+        model (nn.Module): The defined network. Default: None.
+        tile (int): As too large images result in the out of GPU memory issue, so this tile option will first crop
+            input images into tiles, and then process each of them. Finally, they will be merged into one image.
+            0 denotes for do not use tile. Default: 0.
+        tile_pad (int): The pad size for each tile, to remove border artifacts. Default: 10.
+        pre_pad (int): Pad the input images to avoid border artifacts. Default: 10.
+        half (float): Whether to use half precision during inference. Default: False.
+    """
+
+    def __init__(self,
+                 scale,
+                 model_path,
+                 dni_weight=None,
+                 model=None,
+                 tile=0,
+                 tile_pad=10,
+                 pre_pad=10,
+                 half=False,
+                 device=None,
+                 gpu_id=None):
+        self.scale = scale
+        self.tile_size = tile
+        self.tile_pad = tile_pad
+        self.pre_pad = pre_pad
+        self.mod_scale = None
+        self.half = half
+
+        # initialize model
+        if gpu_id:
+            self.device = torch.device(
+                f'cuda:{gpu_id}' if torch.cuda.is_available() else 'cpu') if device is None else device
+        else:
+            self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') if device is None else device
+
+        if isinstance(model_path, list):
+            # dni
+            assert len(model_path) == len(dni_weight), 'model_path and dni_weight should have the save length.'
+            loadnet = self.dni(model_path[0], model_path[1], dni_weight)
+        else:
+            # if the model_path starts with https, it will first download models to the folder: weights
+            if model_path.startswith('https://'):
+                model_path = load_file_from_url(
+                    url=model_path, model_dir=os.path.join(ROOT_DIR, 'weights'), progress=True, file_name=None)
+            loadnet = torch.load(model_path, map_location=torch.device('cpu'))
+
+        # prefer to use params_ema
+        if 'params_ema' in loadnet:
+            keyname = 'params_ema'
+        else:
+            keyname = 'params'
+        model.load_state_dict(loadnet[keyname], strict=True)
+
+        model.eval()
+        self.model = model.to(self.device)
+        if self.half:
+            self.model = self.model.half()
+
+    def dni(self, net_a, net_b, dni_weight, key='params', loc='cpu'):
+        """Deep network interpolation.
+
+        ``Paper: Deep Network Interpolation for Continuous Imagery Effect Transition``
+        """
+        net_a = torch.load(net_a, map_location=torch.device(loc))
+        net_b = torch.load(net_b, map_location=torch.device(loc))
+        for k, v_a in net_a[key].items():
+            net_a[key][k] = dni_weight[0] * v_a + dni_weight[1] * net_b[key][k]
+        return net_a
+
+    def pre_process(self, img):
+        """Pre-process, such as pre-pad and mod pad, so that the images can be divisible
+        """
+        img = torch.from_numpy(np.transpose(img, (2, 0, 1))).float()
+        self.img = img.unsqueeze(0).to(self.device)
+        if self.half:
+            self.img = self.img.half()
+
+        # pre_pad
+        if self.pre_pad != 0:
+            self.img = F.pad(self.img, (0, self.pre_pad, 0, self.pre_pad), 'reflect')
+        # mod pad for divisible borders
+        if self.scale == 2:
+            self.mod_scale = 2
+        elif self.scale == 1:
+            self.mod_scale = 4
+        if self.mod_scale is not None:
+            self.mod_pad_h, self.mod_pad_w = 0, 0
+            _, _, h, w = self.img.size()
+            if (h % self.mod_scale != 0):
+                self.mod_pad_h = (self.mod_scale - h % self.mod_scale)
+            if (w % self.mod_scale != 0):
+                self.mod_pad_w = (self.mod_scale - w % self.mod_scale)
+            self.img = F.pad(self.img, (0, self.mod_pad_w, 0, self.mod_pad_h), 'reflect')
+
+    def process(self):
+        # model inference
+        self.output = self.model(self.img)
+
+    def tile_process(self):
+        """It will first crop input images to tiles, and then process each tile.
+        Finally, all the processed tiles are merged into one images.
+
+        Modified from: https://github.com/ata4/esrgan-launcher
+        """
+        batch, channel, height, width = self.img.shape
+        output_height = height * self.scale
+        output_width = width * self.scale
+        output_shape = (batch, channel, output_height, output_width)
+
+        # start with black image
+        self.output = self.img.new_zeros(output_shape)
+        tiles_x = math.ceil(width / self.tile_size)
+        tiles_y = math.ceil(height / self.tile_size)
+
+        # loop over all tiles
+        for y in range(tiles_y):
+            for x in range(tiles_x):
+                # extract tile from input image
+                ofs_x = x * self.tile_size
+                ofs_y = y * self.tile_size
+                # input tile area on total image
+                input_start_x = ofs_x
+                input_end_x = min(ofs_x + self.tile_size, width)
+                input_start_y = ofs_y
+                input_end_y = min(ofs_y + self.tile_size, height)
+
+                # input tile area on total image with padding
+                input_start_x_pad = max(input_start_x - self.tile_pad, 0)
+                input_end_x_pad = min(input_end_x + self.tile_pad, width)
+                input_start_y_pad = max(input_start_y - self.tile_pad, 0)
+                input_end_y_pad = min(input_end_y + self.tile_pad, height)
+
+                # input tile dimensions
+                input_tile_width = input_end_x - input_start_x
+                input_tile_height = input_end_y - input_start_y
+                tile_idx = y * tiles_x + x + 1
+                input_tile = self.img[:, :, input_start_y_pad:input_end_y_pad, input_start_x_pad:input_end_x_pad]
+
+                # upscale tile
+                try:
+                    with torch.no_grad():
+                        output_tile = self.model(input_tile)
+                except RuntimeError as error:
+                    print('Error', error)
+                print(f'\tTile {tile_idx}/{tiles_x * tiles_y}')
+
+                # output tile area on total image
+                output_start_x = input_start_x * self.scale
+                output_end_x = input_end_x * self.scale
+                output_start_y = input_start_y * self.scale
+                output_end_y = input_end_y * self.scale
+
+                # output tile area without padding
+                output_start_x_tile = (input_start_x - input_start_x_pad) * self.scale
+                output_end_x_tile = output_start_x_tile + input_tile_width * self.scale
+                output_start_y_tile = (input_start_y - input_start_y_pad) * self.scale
+                output_end_y_tile = output_start_y_tile + input_tile_height * self.scale
+
+                # put tile into output image
+                self.output[:, :, output_start_y:output_end_y,
+                            output_start_x:output_end_x] = output_tile[:, :, output_start_y_tile:output_end_y_tile,
+                                                                       output_start_x_tile:output_end_x_tile]
+
+    def post_process(self):
+        # remove extra pad
+        if self.mod_scale is not None:
+            _, _, h, w = self.output.size()
+            self.output = self.output[:, :, 0:h - self.mod_pad_h * self.scale, 0:w - self.mod_pad_w * self.scale]
+        # remove prepad
+        if self.pre_pad != 0:
+            _, _, h, w = self.output.size()
+            self.output = self.output[:, :, 0:h - self.pre_pad * self.scale, 0:w - self.pre_pad * self.scale]
+        return self.output
+
+    @torch.no_grad()
+    def enhance(self, img, outscale=None, alpha_upsampler='realesrgan'):
+        h_input, w_input = img.shape[0:2]
+        # img: numpy
+        img = img.astype(np.float32)
+        if np.max(img) > 256:  # 16-bit image
+            max_range = 65535
+            print('\tInput is a 16-bit image')
+        else:
+            max_range = 255
+        img = img / max_range
+        if len(img.shape) == 2:  # gray image
+            img_mode = 'L'
+            img = cv2.cvtColor(img, cv2.COLOR_GRAY2RGB)
+        elif img.shape[2] == 4:  # RGBA image with alpha channel
+            img_mode = 'RGBA'
+            alpha = img[:, :, 3]
+            img = img[:, :, 0:3]
+            img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+            if alpha_upsampler == 'realesrgan':
+                alpha = cv2.cvtColor(alpha, cv2.COLOR_GRAY2RGB)
+        else:
+            img_mode = 'RGB'
+            img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+
+        # ------------------- process image (without the alpha channel) ------------------- #
+        self.pre_process(img)
+        if self.tile_size > 0:
+            self.tile_process()
+        else:
+            self.process()
+        output_img = self.post_process()
+        output_img = output_img.data.squeeze().float().cpu().clamp_(0, 1).numpy()
+        output_img = np.transpose(output_img[[2, 1, 0], :, :], (1, 2, 0))
+        if img_mode == 'L':
+            output_img = cv2.cvtColor(output_img, cv2.COLOR_BGR2GRAY)
+
+        # ------------------- process the alpha channel if necessary ------------------- #
+        if img_mode == 'RGBA':
+            if alpha_upsampler == 'realesrgan':
+                self.pre_process(alpha)
+                if self.tile_size > 0:
+                    self.tile_process()
+                else:
+                    self.process()
+                output_alpha = self.post_process()
+                output_alpha = output_alpha.data.squeeze().float().cpu().clamp_(0, 1).numpy()
+                output_alpha = np.transpose(output_alpha[[2, 1, 0], :, :], (1, 2, 0))
+                output_alpha = cv2.cvtColor(output_alpha, cv2.COLOR_BGR2GRAY)
+            else:  # use the cv2 resize for alpha channel
+                h, w = alpha.shape[0:2]
+                output_alpha = cv2.resize(alpha, (w * self.scale, h * self.scale), interpolation=cv2.INTER_LINEAR)
+
+            # merge the alpha channel
+            output_img = cv2.cvtColor(output_img, cv2.COLOR_BGR2BGRA)
+            output_img[:, :, 3] = output_alpha
+
+        # ------------------------------ return ------------------------------ #
+        if max_range == 65535:  # 16-bit image
+            output = (output_img * 65535.0).round().astype(np.uint16)
+        else:
+            output = (output_img * 255.0).round().astype(np.uint8)
+
+        if outscale is not None and outscale != float(self.scale):
+            output = cv2.resize(
+                output, (
+                    int(w_input * outscale),
+                    int(h_input * outscale),
+                ), interpolation=cv2.INTER_LANCZOS4)
+
+        return output, img_mode
+
+
+class PrefetchReader(threading.Thread):
+    """Prefetch images.
+
+    Args:
+        img_list (list[str]): A image list of image paths to be read.
+        num_prefetch_queue (int): Number of prefetch queue.
+    """
+
+    def __init__(self, img_list, num_prefetch_queue):
+        super().__init__()
+        self.que = queue.Queue(num_prefetch_queue)
+        self.img_list = img_list
+
+    def run(self):
+        for img_path in self.img_list:
+            img = cv2.imread(img_path, cv2.IMREAD_UNCHANGED)
+            self.que.put(img)
+
+        self.que.put(None)
+
+    def __next__(self):
+        next_item = self.que.get()
+        if next_item is None:
+            raise StopIteration
+        return next_item
+
+    def __iter__(self):
+        return self
+
+
+class IOConsumer(threading.Thread):
+
+    def __init__(self, opt, que, qid):
+        super().__init__()
+        self._queue = que
+        self.qid = qid
+        self.opt = opt
+
+    def run(self):
+        while True:
+            msg = self._queue.get()
+            if isinstance(msg, str) and msg == 'quit':
+                break
+
+            output = msg['output']
+            save_path = msg['save_path']
+            cv2.imwrite(save_path, output)
+        print(f'IO worker {self.qid} is done.')

Real-ESRGAN/realesrgan/version.py

@@ -0,0 +1,5 @@
+# GENERATED VERSION FILE
+# TIME: Mon Apr 14 13:27:54 2025
+__version__ = '0.3.0'
+__gitsha__ = 'a4abfb2'
+version_info = (0, 3, 0)

Real-ESRGAN/scripts/extract_subimages.py

@@ -0,0 +1,135 @@
+import argparse
+import cv2
+import numpy as np
+import os
+import sys
+from basicsr.utils import scandir
+from multiprocessing import Pool
+from os import path as osp
+from tqdm import tqdm
+
+
+def main(args):
+    """A multi-thread tool to crop large images to sub-images for faster IO.
+
+    opt (dict): Configuration dict. It contains:
+        n_thread (int): Thread number.
+        compression_level (int):  CV_IMWRITE_PNG_COMPRESSION from 0 to 9. A higher value means a smaller size
+            and longer compression time. Use 0 for faster CPU decompression. Default: 3, same in cv2.
+        input_folder (str): Path to the input folder.
+        save_folder (str): Path to save folder.
+        crop_size (int): Crop size.
+        step (int): Step for overlapped sliding window.
+        thresh_size (int): Threshold size. Patches whose size is lower than thresh_size will be dropped.
+
+    Usage:
+        For each folder, run this script.
+        Typically, there are GT folder and LQ folder to be processed for DIV2K dataset.
+        After process, each sub_folder should have the same number of subimages.
+        Remember to modify opt configurations according to your settings.
+    """
+
+    opt = {}
+    opt['n_thread'] = args.n_thread
+    opt['compression_level'] = args.compression_level
+    opt['input_folder'] = args.input
+    opt['save_folder'] = args.output
+    opt['crop_size'] = args.crop_size
+    opt['step'] = args.step
+    opt['thresh_size'] = args.thresh_size
+    extract_subimages(opt)
+
+
+def extract_subimages(opt):
+    """Crop images to subimages.
+
+    Args:
+        opt (dict): Configuration dict. It contains:
+            input_folder (str): Path to the input folder.
+            save_folder (str): Path to save folder.
+            n_thread (int): Thread number.
+    """
+    input_folder = opt['input_folder']
+    save_folder = opt['save_folder']
+    if not osp.exists(save_folder):
+        os.makedirs(save_folder)
+        print(f'mkdir {save_folder} ...')
+    else:
+        print(f'Folder {save_folder} already exists. Exit.')
+        sys.exit(1)
+
+    # scan all images
+    img_list = list(scandir(input_folder, full_path=True))
+
+    pbar = tqdm(total=len(img_list), unit='image', desc='Extract')
+    pool = Pool(opt['n_thread'])
+    for path in img_list:
+        pool.apply_async(worker, args=(path, opt), callback=lambda arg: pbar.update(1))
+    pool.close()
+    pool.join()
+    pbar.close()
+    print('All processes done.')
+
+
+def worker(path, opt):
+    """Worker for each process.
+
+    Args:
+        path (str): Image path.
+        opt (dict): Configuration dict. It contains:
+            crop_size (int): Crop size.
+            step (int): Step for overlapped sliding window.
+            thresh_size (int): Threshold size. Patches whose size is lower than thresh_size will be dropped.
+            save_folder (str): Path to save folder.
+            compression_level (int): for cv2.IMWRITE_PNG_COMPRESSION.
+
+    Returns:
+        process_info (str): Process information displayed in progress bar.
+    """
+    crop_size = opt['crop_size']
+    step = opt['step']
+    thresh_size = opt['thresh_size']
+    img_name, extension = osp.splitext(osp.basename(path))
+
+    # remove the x2, x3, x4 and x8 in the filename for DIV2K
+    img_name = img_name.replace('x2', '').replace('x3', '').replace('x4', '').replace('x8', '')
+
+    img = cv2.imread(path, cv2.IMREAD_UNCHANGED)
+
+    h, w = img.shape[0:2]
+    h_space = np.arange(0, h - crop_size + 1, step)
+    if h - (h_space[-1] + crop_size) > thresh_size:
+        h_space = np.append(h_space, h - crop_size)
+    w_space = np.arange(0, w - crop_size + 1, step)
+    if w - (w_space[-1] + crop_size) > thresh_size:
+        w_space = np.append(w_space, w - crop_size)
+
+    index = 0
+    for x in h_space:
+        for y in w_space:
+            index += 1
+            cropped_img = img[x:x + crop_size, y:y + crop_size, ...]
+            cropped_img = np.ascontiguousarray(cropped_img)
+            cv2.imwrite(
+                osp.join(opt['save_folder'], f'{img_name}_s{index:03d}{extension}'), cropped_img,
+                [cv2.IMWRITE_PNG_COMPRESSION, opt['compression_level']])
+    process_info = f'Processing {img_name} ...'
+    return process_info
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--input', type=str, default='datasets/DF2K/DF2K_HR', help='Input folder')
+    parser.add_argument('--output', type=str, default='datasets/DF2K/DF2K_HR_sub', help='Output folder')
+    parser.add_argument('--crop_size', type=int, default=480, help='Crop size')
+    parser.add_argument('--step', type=int, default=240, help='Step for overlapped sliding window')
+    parser.add_argument(
+        '--thresh_size',
+        type=int,
+        default=0,
+        help='Threshold size. Patches whose size is lower than thresh_size will be dropped.')
+    parser.add_argument('--n_thread', type=int, default=20, help='Thread number.')
+    parser.add_argument('--compression_level', type=int, default=3, help='Compression level')
+    args = parser.parse_args()
+
+    main(args)

Real-ESRGAN/scripts/generate_meta_info.py

@@ -0,0 +1,58 @@
+import argparse
+import cv2
+import glob
+import os
+
+
+def main(args):
+    txt_file = open(args.meta_info, 'w')
+    for folder, root in zip(args.input, args.root):
+        img_paths = sorted(glob.glob(os.path.join(folder, '*')))
+        for img_path in img_paths:
+            status = True
+            if args.check:
+                # read the image once for check, as some images may have errors
+                try:
+                    img = cv2.imread(img_path)
+                except (IOError, OSError) as error:
+                    print(f'Read {img_path} error: {error}')
+                    status = False
+                if img is None:
+                    status = False
+                    print(f'Img is None: {img_path}')
+            if status:
+                # get the relative path
+                img_name = os.path.relpath(img_path, root)
+                print(img_name)
+                txt_file.write(f'{img_name}\n')
+
+
+if __name__ == '__main__':
+    """Generate meta info (txt file) for only Ground-Truth images.
+
+    It can also generate meta info from several folders into one txt file.
+    """
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        '--input',
+        nargs='+',
+        default=['datasets/DF2K/DF2K_HR', 'datasets/DF2K/DF2K_multiscale'],
+        help='Input folder, can be a list')
+    parser.add_argument(
+        '--root',
+        nargs='+',
+        default=['datasets/DF2K', 'datasets/DF2K'],
+        help='Folder root, should have the length as input folders')
+    parser.add_argument(
+        '--meta_info',
+        type=str,
+        default='datasets/DF2K/meta_info/meta_info_DF2Kmultiscale.txt',
+        help='txt path for meta info')
+    parser.add_argument('--check', action='store_true', help='Read image to check whether it is ok')
+    args = parser.parse_args()
+
+    assert len(args.input) == len(args.root), ('Input folder and folder root should have the same length, but got '
+                                               f'{len(args.input)} and {len(args.root)}.')
+    os.makedirs(os.path.dirname(args.meta_info), exist_ok=True)
+
+    main(args)

Real-ESRGAN/scripts/generate_meta_info_pairdata.py

@@ -0,0 +1,49 @@
+import argparse
+import glob
+import os
+
+
+def main(args):
+    txt_file = open(args.meta_info, 'w')
+    # sca images
+    img_paths_gt = sorted(glob.glob(os.path.join(args.input[0], '*')))
+    img_paths_lq = sorted(glob.glob(os.path.join(args.input[1], '*')))
+
+    assert len(img_paths_gt) == len(img_paths_lq), ('GT folder and LQ folder should have the same length, but got '
+                                                    f'{len(img_paths_gt)} and {len(img_paths_lq)}.')
+
+    for img_path_gt, img_path_lq in zip(img_paths_gt, img_paths_lq):
+        # get the relative paths
+        img_name_gt = os.path.relpath(img_path_gt, args.root[0])
+        img_name_lq = os.path.relpath(img_path_lq, args.root[1])
+        print(f'{img_name_gt}, {img_name_lq}')
+        txt_file.write(f'{img_name_gt}, {img_name_lq}\n')
+
+
+if __name__ == '__main__':
+    """This script is used to generate meta info (txt file) for paired images.
+    """
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        '--input',
+        nargs='+',
+        default=['datasets/DF2K/DIV2K_train_HR_sub', 'datasets/DF2K/DIV2K_train_LR_bicubic_X4_sub'],
+        help='Input folder, should be [gt_folder, lq_folder]')
+    parser.add_argument('--root', nargs='+', default=[None, None], help='Folder root, will use the ')
+    parser.add_argument(
+        '--meta_info',
+        type=str,
+        default='datasets/DF2K/meta_info/meta_info_DIV2K_sub_pair.txt',
+        help='txt path for meta info')
+    args = parser.parse_args()
+
+    assert len(args.input) == 2, 'Input folder should have two elements: gt folder and lq folder'
+    assert len(args.root) == 2, 'Root path should have two elements: root for gt folder and lq folder'
+    os.makedirs(os.path.dirname(args.meta_info), exist_ok=True)
+    for i in range(2):
+        if args.input[i].endswith('/'):
+            args.input[i] = args.input[i][:-1]
+        if args.root[i] is None:
+            args.root[i] = os.path.dirname(args.input[i])
+
+    main(args)

Real-ESRGAN/scripts/generate_multiscale_DF2K.py

@@ -0,0 +1,48 @@
+import argparse
+import glob
+import os
+from PIL import Image
+
+
+def main(args):
+    # For DF2K, we consider the following three scales,
+    # and the smallest image whose shortest edge is 400
+    scale_list = [0.75, 0.5, 1 / 3]
+    shortest_edge = 400
+
+    path_list = sorted(glob.glob(os.path.join(args.input, '*')))
+    for path in path_list:
+        print(path)
+        basename = os.path.splitext(os.path.basename(path))[0]
+
+        img = Image.open(path)
+        width, height = img.size
+        for idx, scale in enumerate(scale_list):
+            print(f'\t{scale:.2f}')
+            rlt = img.resize((int(width * scale), int(height * scale)), resample=Image.LANCZOS)
+            rlt.save(os.path.join(args.output, f'{basename}T{idx}.png'))
+
+        # save the smallest image which the shortest edge is 400
+        if width < height:
+            ratio = height / width
+            width = shortest_edge
+            height = int(width * ratio)
+        else:
+            ratio = width / height
+            height = shortest_edge
+            width = int(height * ratio)
+        rlt = img.resize((int(width), int(height)), resample=Image.LANCZOS)
+        rlt.save(os.path.join(args.output, f'{basename}T{idx+1}.png'))
+
+
+if __name__ == '__main__':
+    """Generate multi-scale versions for GT images with LANCZOS resampling.
+    It is now used for DF2K dataset (DIV2K + Flickr 2K)
+    """
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--input', type=str, default='datasets/DF2K/DF2K_HR', help='Input folder')
+    parser.add_argument('--output', type=str, default='datasets/DF2K/DF2K_multiscale', help='Output folder')
+    args = parser.parse_args()
+
+    os.makedirs(args.output, exist_ok=True)
+    main(args)

Real-ESRGAN/scripts/pytorch2onnx.py

@@ -0,0 +1,36 @@
+import argparse
+import torch
+import torch.onnx
+from basicsr.archs.rrdbnet_arch import RRDBNet
+
+
+def main(args):
+    # An instance of the model
+    model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=4)
+    if args.params:
+        keyname = 'params'
+    else:
+        keyname = 'params_ema'
+    model.load_state_dict(torch.load(args.input)[keyname])
+    # set the train mode to false since we will only run the forward pass.
+    model.train(False)
+    model.cpu().eval()
+
+    # An example input
+    x = torch.rand(1, 3, 64, 64)
+    # Export the model
+    with torch.no_grad():
+        torch_out = torch.onnx._export(model, x, args.output, opset_version=11, export_params=True)
+    print(torch_out.shape)
+
+
+if __name__ == '__main__':
+    """Convert pytorch model to onnx models"""
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        '--input', type=str, default='experiments/pretrained_models/RealESRGAN_x4plus.pth', help='Input model path')
+    parser.add_argument('--output', type=str, default='realesrgan-x4.onnx', help='Output onnx path')
+    parser.add_argument('--params', action='store_false', help='Use params instead of params_ema')
+    args = parser.parse_args()
+
+    main(args)

Real-ESRGAN/setup.py

@@ -0,0 +1,107 @@
+#!/usr/bin/env python
+
+from setuptools import find_packages, setup
+
+import os
+import subprocess
+import time
+
+version_file = 'realesrgan/version.py'
+
+
+def readme():
+    with open('README.md', encoding='utf-8') as f:
+        content = f.read()
+    return content
+
+
+def get_git_hash():
+
+    def _minimal_ext_cmd(cmd):
+        # construct minimal environment
+        env = {}
+        for k in ['SYSTEMROOT', 'PATH', 'HOME']:
+            v = os.environ.get(k)
+            if v is not None:
+                env[k] = v
+        # LANGUAGE is used on win32
+        env['LANGUAGE'] = 'C'
+        env['LANG'] = 'C'
+        env['LC_ALL'] = 'C'
+        out = subprocess.Popen(cmd, stdout=subprocess.PIPE, env=env).communicate()[0]
+        return out
+
+    try:
+        out = _minimal_ext_cmd(['git', 'rev-parse', 'HEAD'])
+        sha = out.strip().decode('ascii')
+    except OSError:
+        sha = 'unknown'
+
+    return sha
+
+
+def get_hash():
+    if os.path.exists('.git'):
+        sha = get_git_hash()[:7]
+    else:
+        sha = 'unknown'
+
+    return sha
+
+
+def write_version_py():
+    content = """# GENERATED VERSION FILE
+# TIME: {}
+__version__ = '{}'
+__gitsha__ = '{}'
+version_info = ({})
+"""
+    sha = get_hash()
+    with open('VERSION', 'r') as f:
+        SHORT_VERSION = f.read().strip()
+    VERSION_INFO = ', '.join([x if x.isdigit() else f'"{x}"' for x in SHORT_VERSION.split('.')])
+
+    version_file_str = content.format(time.asctime(), SHORT_VERSION, sha, VERSION_INFO)
+    with open(version_file, 'w') as f:
+        f.write(version_file_str)
+
+
+def get_version():
+    with open(version_file, 'r') as f:
+        exec(compile(f.read(), version_file, 'exec'))
+    return locals()['__version__']
+
+
+def get_requirements(filename='requirements.txt'):
+    here = os.path.dirname(os.path.realpath(__file__))
+    with open(os.path.join(here, filename), 'r') as f:
+        requires = [line.replace('\n', '') for line in f.readlines()]
+    return requires
+
+
+if __name__ == '__main__':
+    write_version_py()
+    setup(
+        name='realesrgan',
+        version=get_version(),
+        description='Real-ESRGAN aims at developing Practical Algorithms for General Image Restoration',
+        long_description=readme(),
+        long_description_content_type='text/markdown',
+        author='Xintao Wang',
+        author_email='xintao.wang@outlook.com',
+        keywords='computer vision, pytorch, image restoration, super-resolution, esrgan, real-esrgan',
+        url='https://github.com/xinntao/Real-ESRGAN',
+        include_package_data=True,
+        packages=find_packages(exclude=('options', 'datasets', 'experiments', 'results', 'tb_logger', 'wandb')),
+        classifiers=[
+            'Development Status :: 4 - Beta',
+            'License :: OSI Approved :: Apache Software License',
+            'Operating System :: OS Independent',
+            'Programming Language :: Python :: 3',
+            'Programming Language :: Python :: 3.7',
+            'Programming Language :: Python :: 3.8',
+        ],
+        license='BSD-3-Clause License',
+        setup_requires=['cython', 'numpy'],
+        install_requires=get_requirements(),
+        zip_safe=False)

Real-ESRGAN/weights/README.md

@@ -0,0 +1,3 @@
+# Weights
+
+Put the downloaded weights to this folder.

Real-ESRGAN/weights/RealESRGAN_x2plus.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49fafd45f8fd7aa8d31ab2a22d14d91b536c34494a5cfe31eb5d89c2fa266abb
+size 67061725

Real-ESRGAN/weights/RealESRGAN_x4plus.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fa0d38905f75ac06eb49a7951b426670021be3018265fd191d2125df9d682f1
+size 67040989

api.py

@@ -0,0 +1,377 @@
+import os
+import uuid
+import gc
+import subprocess
+import sys
+import traceback
+import shutil
+import logging
+from typing import Optional, List
+from pathlib import Path
+
+from fastapi import FastAPI, File, UploadFile, HTTPException, BackgroundTasks, Form
+from fastapi.responses import FileResponse
+from fastapi.middleware.cors import CORSMiddleware
+import uvicorn
+import psutil
+
+# --- Configuration ---
+SCRIPT_DIR = Path(__file__).parent.resolve()
+REAL_ESRGAN_DIR = SCRIPT_DIR / "Real-ESRGAN"
+INFERENCE_SCRIPT = REAL_ESRGAN_DIR / "inference_realesrgan.py"
+MODEL_DIR = REAL_ESRGAN_DIR / "weights"
+INPUT_DIR = SCRIPT_DIR / "api_inputs"
+OUTPUT_DIR = SCRIPT_DIR / "api_outputs"
+API_PORT = 8000
+LOG_FILE = SCRIPT_DIR / "api.log"
+
+# --- Setup Logging ---
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
+    handlers=[
+        logging.FileHandler(LOG_FILE),
+        logging.StreamHandler(sys.stdout) # Also print logs to console
+    ]
+)
+logger = logging.getLogger(__name__)
+
+# --- Create Directories ---
+INPUT_DIR.mkdir(exist_ok=True)
+OUTPUT_DIR.mkdir(exist_ok=True)
+
+# --- FastAPI App Initialization ---
+app = FastAPI(
+    title="Image Enhancer API",
+    description="API for enhancing images.",
+    version="1.0.0"
+)
+
+# --- CORS Middleware ---
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"], # Allow all origins for simplicity, adjust in production
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# --- Global State ---
+processing_lock = False
+available_models = []
+DEFAULT_MODEL_PREFERENCE = "RealESRGAN_x4plus" # Preferred default
+
+# Define allowed values for API input validation
+AVAILABLE_MODELS_API = ["RealESRGAN_x4plus", "RealESRGAN_x2plus"]
+ALLOWED_SCALES_API = [1.0, 2.0, 4.0, 8.0]
+DEFAULT_MODEL_API = "RealESRGAN_x4plus"
+DEFAULT_SCALE_API = 4.0
+DEFAULT_TILE_SIZE = 400 # Default tile size to use on memory error retry
+
+def update_available_models():
+    """Scans the model directory and updates the list of available models."""
+    global available_models
+    try:
+        models = [f.stem for f in MODEL_DIR.glob("*.pth")]
+        if not models:
+            logger.warning(f"No model files (.pth) found in {MODEL_DIR}")
+        available_models = sorted(models)
+        logger.info(f"Available models updated: {available_models}")
+    except Exception as e:
+        logger.error(f"Error scanning model directory {MODEL_DIR}: {e}")
+        available_models = []
+
+# Initialize models on startup
+update_available_models()
+
+
+# --- Helper Functions ---
+def release_lock():
+    """Releases the processing lock."""
+    global processing_lock
+    processing_lock = False
+    logger.info("Processing lock released.")
+
+# --- API Endpoints ---
+@app.get("/")
+async def root():
+    """Root endpoint providing basic API information."""
+    return {"message": "Image Enhancer API is running"}
+
+@app.get("/models/", response_model=List[str])
+async def get_models():
+    """Returns a list of available Real-ESRGAN models."""
+    if not available_models:
+        update_available_models() # Attempt to rescan if list is empty
+    if not available_models:
+         raise HTTPException(status_code=404, detail=f"No models found in {MODEL_DIR}")
+    return available_models
+
+@app.post("/enhance/", response_class=FileResponse)
+async def enhance_image(
+    background_tasks: BackgroundTasks,
+    file: UploadFile = File(...),
+    model_name: Optional[str] = Form(DEFAULT_MODEL_API),
+    outscale: float = Form(DEFAULT_SCALE_API),
+    face_enhance: bool = Form(False),
+    fp32: bool = Form(False),
+    tile: Optional[int] = Form(0)
+):
+    """
+    Enhances an uploaded image using Real-ESRGAN.
+    Automatically retries with tiling if an out-of-memory error is detected.
+    """
+    global processing_lock
+    temp_input_path = None
+    temp_output_dir_for_request = None
+    temp_input_dir_for_request = None # Added for consistency
+
+    # --- Request Handling ---
+    request_id = uuid.uuid4().hex
+    logger.info(f"Received enhancement request ID: {request_id}")
+
+    # Check processing lock
+    if processing_lock:
+        logger.warning(f"Request {request_id}: Server busy, denying request.")
+        raise HTTPException(
+            status_code=429,
+            detail="Server is busy processing another image. Please try again shortly."
+        )
+    processing_lock = True
+    logger.info(f"Request {request_id}: Processing lock acquired.")
+
+    # --- Input Validation ---
+    # Validate model name against allowed list
+    if model_name not in AVAILABLE_MODELS_API:
+        logger.warning(f"Request {request_id}: Invalid model_name specified: '{model_name}'. Allowed: {AVAILABLE_MODELS_API}")
+        release_lock()
+        raise HTTPException(
+            status_code=400,
+            detail=f"Invalid model name '{model_name}'. Allowed values: {AVAILABLE_MODELS_API}"
+        )
+
+    # Validate scale against allowed list
+    if outscale not in ALLOWED_SCALES_API:
+        logger.warning(f"Request {request_id}: Invalid outscale specified: '{outscale}'. Allowed: {ALLOWED_SCALES_API}")
+        release_lock()
+        raise HTTPException(
+            status_code=400,
+            detail=f"Invalid scale value '{outscale}'. Allowed values: {ALLOWED_SCALES_API}"
+        )
+
+    # Validate file type
+    if not file.content_type or not file.content_type.startswith("image/"):
+        logger.warning(f"Request {request_id}: Invalid file type uploaded: {file.content_type}")
+        release_lock()
+        raise HTTPException(status_code=400, detail="Invalid file type. Please upload an image.")
+
+    # --- Model Existence Check ---
+    # Check if the validated model actually exists in the scanned directory
+    if model_name not in available_models:
+        logger.error(f"Request {request_id}: Model '{model_name}' is allowed but not found in {MODEL_DIR}. Scanned models: {available_models}")
+        update_available_models() # Try rescanning
+        if model_name not in available_models:
+             release_lock()
+             raise HTTPException(
+                 status_code=500,
+                 detail=f"Model '{model_name}' not found on server, even though it's an allowed option. Please check server configuration."
+             )
+
+    final_model_name = model_name # Use the validated model name
+    logger.info(f"Request {request_id}: Using validated model: {final_model_name}, scale: {outscale}")
+
+    try:
+        # --- File Handling ---
+        # Create unique temporary paths for this request
+        input_suffix = Path(file.filename).suffix if file.filename else '.png'
+        # Use original filename for input file within its own request dir
+        temp_input_filename = Path(file.filename).name if file.filename else f"input_{request_id}{input_suffix}"
+
+        # Input directory for this specific request
+        temp_input_dir_for_request = INPUT_DIR / request_id
+        temp_input_dir_for_request.mkdir(exist_ok=True)
+        temp_input_path = temp_input_dir_for_request / temp_input_filename
+
+        # Output directory for this specific request's results
+        temp_output_dir_for_request = OUTPUT_DIR / request_id
+        temp_output_dir_for_request.mkdir(exist_ok=True)
+
+        # Save uploaded file to its request-specific input dir
+        try:
+            logger.info(f"Request {request_id}: Saving uploaded file to {temp_input_path}")
+            contents = await file.read()
+            with open(temp_input_path, "wb") as buffer:
+                buffer.write(contents)
+            logger.info(f"Request {request_id}: Uploaded file saved successfully.")
+        except Exception as e:
+            logger.error(f"Request {request_id}: Failed to save uploaded file: {e}")
+            raise HTTPException(status_code=500, detail="Failed to save uploaded file.")
+        finally:
+            await file.close() # Ensure file handle is closed
+
+        # --- Inference Execution ---
+        # Construct command (base_cmd now uses temp_input_path which includes the subdir)
+        base_cmd = [
+            sys.executable, str(INFERENCE_SCRIPT),
+            "-i", str(temp_input_path),
+            "-o", str(temp_output_dir_for_request),
+            "-n", final_model_name,
+            "-s", str(outscale),
+        ]
+        if face_enhance:
+            base_cmd.append("--face_enhance")
+        if fp32:
+            base_cmd.append("--fp32")
+        # Add tile param only if explicitly provided (> 0) or during retry
+        if tile > 0:
+             base_cmd.extend(["-t", str(tile)])
+
+        logger.info(f"Request {request_id}: Preparing initial inference command...")
+
+        # Execute the script - Attempt 1 (No Tile unless specified)
+        try:
+            logger.info(f"Request {request_id}: Running inference (Attempt 1): {' '.join(base_cmd)}")
+            process = subprocess.run(
+                base_cmd,
+                capture_output=True,
+                text=True,
+                check=True,
+                cwd=REAL_ESRGAN_DIR
+            )
+            logger.info(f"Request {request_id}: Inference script (Attempt 1) stdout:{process.stdout}")
+            if process.stderr:
+                 logger.warning(f"Request {request_id}: Inference script (Attempt 1) stderr:{process.stderr}")
+
+        except (subprocess.CalledProcessError, RuntimeError) as e:
+            error_output = ""
+            if isinstance(e, subprocess.CalledProcessError):
+                error_output = e.stderr
+                logger.error(f"Request {request_id}: Inference script failed (Attempt 1) with exit code {e.returncode}")
+                logger.error(f"Request {request_id}: Stdout: {e.stdout}")
+                logger.error(f"Request {request_id}: Stderr: {e.stderr}")
+            else: # Handle RuntimeError which might be raised by realesrgan directly
+                error_output = str(e)
+                logger.error(f"Request {request_id}: Inference script raised RuntimeError (Attempt 1): {e}")
+
+            # Check if it's a memory error and tile wasn't already manually set
+            is_memory_error = "memory" in error_output.lower() or "cuda" in error_output.lower()
+            tile_arg_present = any(arg == "-t" for arg in base_cmd)
+
+            if is_memory_error and not tile_arg_present:
+                logger.warning(f"Request {request_id}: Detected potential memory error. Retrying with tiling (tile_size={DEFAULT_TILE_SIZE})...")
+                # Attempt 2 (With Tile)
+                retry_cmd = base_cmd + ["-t", str(DEFAULT_TILE_SIZE)]
+                try:
+                    logger.info(f"Request {request_id}: Running inference (Attempt 2 - Tiled): {' '.join(retry_cmd)}")
+                    process = subprocess.run(
+                        retry_cmd,
+                        capture_output=True,
+                        text=True,
+                        check=True,
+                        cwd=REAL_ESRGAN_DIR
+                    )
+                    logger.info(f"Request {request_id}: Inference script (Attempt 2 - Tiled) stdout:{process.stdout}")
+                    if process.stderr:
+                        logger.warning(f"Request {request_id}: Inference script (Attempt 2 - Tiled) stderr:{process.stderr}")
+                except (subprocess.CalledProcessError, RuntimeError) as e2:
+                    logger.error(f"Request {request_id}: Inference script failed even on retry with tiling.")
+                    # Log the second error
+                    if isinstance(e2, subprocess.CalledProcessError):
+                         logger.error(f"Request {request_id}: Retry Exit Code: {e2.returncode}, Stderr: {e2.stderr}")
+                         error_output = e2.stderr # Use the error from the retry attempt
+                    else:
+                         logger.error(f"Request {request_id}: Retry RuntimeError: {e2}")
+                         error_output = str(e2)
+                    # Raise original error type but with potentially updated message from retry
+                    raise HTTPException(status_code=500, detail=f"Image enhancement failed, even with tiling: {error_output or 'Unknown error'}")
+            else:
+                # Not a memory error, or tile was already specified - fail normally
+                 raise HTTPException(status_code=500, detail=f"Image enhancement script failed: {error_output or 'Unknown error'}")
+
+        except Exception as e:
+            # Catch any other unexpected errors during subprocess execution
+            logger.error(f"Request {request_id}: Unexpected error executing inference script: {e}")
+            logger.error(traceback.format_exc())
+            raise HTTPException(status_code=500, detail=f"Failed to run enhancement process: {e}")
+
+        # --- Result Handling ---
+        # Find the output file (assumes script outputs one file with '_out' suffix)
+        # The script `inference_realesrgan.py` saves the output as `{basename}_out.{ext}`
+        original_basename = Path(temp_input_filename).stem
+        expected_output_stem = f"{original_basename}_out"
+        output_files = list(temp_output_dir_for_request.glob(f"{expected_output_stem}.*"))
+
+        if not output_files:
+            logger.error(f"Request {request_id}: No output file found in {temp_output_dir_for_request} matching stem {expected_output_stem}")
+            raise HTTPException(status_code=500, detail="Enhancement finished, but output file not found.")
+
+        output_path = output_files[0]
+        output_media_type = f"image/{output_path.suffix.strip('.')}"
+        output_filename = f"enhanced_{Path(file.filename).name}" if file.filename else f"enhanced_{request_id}{output_path.suffix}"
+
+        logger.info(f"Request {request_id}: Enhancement successful. Output: {output_path}")
+
+        # Schedule cleanup task (input file and the whole output dir for this request)
+        # background_tasks.add_task(cleanup_files, [temp_input_path, temp_output_dir_for_request]) # Removed cleanup
+        # Release lock AFTER scheduling cleanup but BEFORE returning response
+        background_tasks.add_task(release_lock)
+
+        # Return the enhanced image file
+        return FileResponse(
+            path=output_path,
+            media_type=output_media_type,
+            filename=output_filename
+        )
+
+    except HTTPException as http_exc:
+         # If an HTTPException occurred (validation, busy, etc.), release lock immediately
+         release_lock()
+         # Re-raise the exception to be handled by FastAPI
+         raise http_exc
+    except Exception as e:
+        error_msg = f"Request {request_id}: Unexpected error during enhancement: {str(e)}"
+        logger.error(error_msg)
+        logger.error(traceback.format_exc())
+        # Ensure cleanup happens even on unexpected errors (Cleanup is removed, but keep release_lock)
+        # We need to potentially clean up the created input directory as well if saving failed
+        # For simplicity now, inputs/outputs persist on errors too, consistent with success path
+        release_lock()
+        raise HTTPException(status_code=500, detail=f"An unexpected error occurred: {str(e)}")
+
+
+@app.get("/status/")
+async def status():
+    """Checks the API status and resource usage."""
+    logger.info("Status check requested.")
+    return {
+        "status": "ok" if not processing_lock else "busy",
+        "processing_active": processing_lock,
+        "available_models": available_models,
+        "memory_usage": {
+            "percent": f"{psutil.virtual_memory().percent}%",
+            "available": f"{psutil.virtual_memory().available / (1024**3):.2f} GB",
+        },
+        "cpu_usage": f"{psutil.cpu_percent()}%",
+        "real_esrgan_dir_exists": REAL_ESRGAN_DIR.exists(),
+        "inference_script_exists": INFERENCE_SCRIPT.exists(),
+        "model_dir_exists": MODEL_DIR.exists(),
+        "input_dir_exists": INPUT_DIR.exists(),
+        "output_dir_exists": OUTPUT_DIR.exists(),
+    }
+
+# --- Server Execution ---
+if __name__ == "__main__":
+    logger.info(f"Starting Image Enhancer API server on port {API_PORT}...")
+    logger.info(f"Real-ESRGAN Directory: {REAL_ESRGAN_DIR}")
+    logger.info(f"Inference Script: {INFERENCE_SCRIPT}")
+    logger.info(f"Model Directory: {MODEL_DIR}")
+    logger.info(f"API Input Directory: {INPUT_DIR}")
+    logger.info(f"API Output Directory: {OUTPUT_DIR}")
+    update_available_models() # Ensure models are listed on startup
+    uvicorn.run(
+        "api:app",
+        host="0.0.0.0",
+        port=API_PORT,
+        reload=False, # Use reload carefully, can cause issues with locking/models
+        log_level="info" # Uvicorn's own log level
+    ) 

app.py

@@ -0,0 +1,273 @@
+import streamlit as st
+import requests
+from PIL import Image
+import io
+import time
+from pathlib import Path
+
+# --- Configuration ---
+API_URL = "http://localhost:8000" # Keep API port
+PAGE_TITLE = "Image Enhancer App"
+PAGE_ICON = "✨"
+# Define fixed model options and default
+AVAILABLE_MODELS = ["RealESRGAN_x4plus", "RealESRGAN_x2plus"]
+DEFAULT_MODEL = "RealESRGAN_x4plus"
+# Define fixed scale options and default
+AVAILABLE_SCALES = [1, 2, 4, 8]
+DEFAULT_SCALE = 4
+
+# --- Initialize Session State ---
+# Use get to avoid errors if keys don't exist yet after a code change/refresh
+st.session_state.setdefault('enhanced_image_data', None)
+st.session_state.setdefault('enhanced_image_caption', None)
+st.session_state.setdefault('download_filename', None)
+st.session_state.setdefault('download_mime', None)
+st.session_state.setdefault('current_file_identifier', None)
+st.session_state.setdefault('error_message', None)
+
+# --- Streamlit Page Setup ---
+st.set_page_config(
+    page_title=PAGE_TITLE,
+    page_icon=PAGE_ICON,
+    layout="wide",
+)
+
+st.title(PAGE_TITLE)
+st.markdown("""
+Enhance your images using the power of AI upscaling.
+Upload an image and choose your enhancement options.
+""")
+
+# --- Helper Functions ---
+# @st.cache_data # Cache the status check result for a short time? Maybe not needed.
+def get_api_status():
+    """Checks the status of the backend API."""
+    try:
+        response = requests.get(f"{API_URL}/status/", timeout=5)
+        if response.status_code == 200:
+            return response.json()
+        else:
+            st.error(f"API Error: Status code {response.status_code}")
+            return None
+    except requests.exceptions.RequestException as e:
+        st.error(f"API Connection Error: {e}")
+        return None
+
+# No longer fetching models from API for selection, using fixed list
+# def get_available_models():
+#     """Fetches the list of available models from the API."""
+#     try:
+#         response = requests.get(f"{API_URL}/models/", timeout=5)
+#         if response.status_code == 200:
+#             return response.json()
+#         else:
+#             st.warning(f"Could not fetch models (Status: {response.status_code}). Using defaults.")
+#             return ["RealESRGAN_x4plus", "RealESRGAN_x2plus"] # Fallback
+#     except requests.exceptions.RequestException as e:
+#         st.warning(f"Could not fetch models (Error: {e}). Using defaults.")
+#         return ["RealESRGAN_x4plus", "RealESRGAN_x2plus"] # Fallback
+
+# --- Sidebar Controls ---
+with st.sidebar:
+    st.header("Enhancement Options")
+
+    # Model selection using fixed list
+    # Ensure default is selected if available, otherwise fallback to first item
+    default_model_index = AVAILABLE_MODELS.index(DEFAULT_MODEL) if DEFAULT_MODEL in AVAILABLE_MODELS else 0
+    selected_model = st.selectbox(
+        "Select Model",
+        AVAILABLE_MODELS,
+        index=default_model_index
+    )
+
+    # Scale factor selection using fixed list
+    default_scale_index = AVAILABLE_SCALES.index(DEFAULT_SCALE) if DEFAULT_SCALE in AVAILABLE_SCALES else 0
+    output_scale = st.selectbox(
+        "Output Scale Factor",
+        AVAILABLE_SCALES,
+        index=default_scale_index
+    )
+    # Remove number input and auto-detect logic
+    # # Determine default scale based on model name if possible
+    # default_scale = 2.0 if selected_model and 'x2' in selected_model else 4.0
+    # output_scale = st.number_input("Output Scale Factor", min_value=1.0, max_value=8.0, value=default_scale, step=0.1)
+
+    # Checkboxes for boolean flags
+    face_enhance = st.checkbox("Enable Face Enhancement (GFPGAN)", value=False)
+    use_fp32 = st.checkbox("Use FP32 Precision (Slower, More Memory)", value=False)
+
+    st.markdown("---")
+    st.header("API Status")
+    if st.button("Check API Status"):
+        status_info = get_api_status()
+        if status_info:
+            status_text = status_info.get("status", "unknown")
+            if status_text == "ok":
+                st.success("✅ API is running and ready.")
+            elif status_text == "busy":
+                st.warning("⏳ API is currently busy processing.")
+            else:
+                st.error(f"❌ API reported status: {status_text}.")
+
+            # Display API-reported models for confirmation
+            api_models = status_info.get("available_models", [])
+            if api_models:
+                st.write(f"**API Models Found:** {', '.join(api_models)}")
+                # Check if selected model is actually available according to API
+                if selected_model not in api_models:
+                    st.warning(f"Selected model '{selected_model}' not found by API!")
+            else:
+                 st.warning("Could not verify available models from API.")
+
+            if "memory_usage" in status_info:
+                st.write(f"**Memory:** {status_info['memory_usage'].get('percent', 'N/A')}")
+        # Error handling is done within get_api_status
+
+# --- Main Area --- 
+
+# File uploader
+uploaded_file = st.file_uploader(
+    "Choose an image to enhance...", type=["jpg", "jpeg", "png", "bmp", "webp"]
+)
+
+if uploaded_file is not None:
+    # Use name + size as a relatively stable identifier across uploads
+    current_file_identifier = f"{uploaded_file.name}-{uploaded_file.size}"
+
+    # --- Reset state if a new file is uploaded ---
+    if current_file_identifier != st.session_state.get('current_file_identifier'):
+        st.session_state.enhanced_image_data = None
+        st.session_state.enhanced_image_caption = None
+        st.session_state.download_filename = None
+        st.session_state.download_mime = None
+        st.session_state.error_message = None # Clear previous errors
+        st.session_state.current_file_identifier = current_file_identifier
+
+    # Display the original image
+    col1, col2 = st.columns(2)
+    with col1:
+        st.subheader("Original Image")
+        try:
+            original_image = Image.open(uploaded_file)
+            st.image(original_image, use_column_width=True, caption="Original")
+
+            # --- Moved Enhance Button Here ---
+            enhance_button_pressed = st.button("Enhance Image ✨")
+
+        except Exception as e:
+            st.error(f"Error loading image: {e}")
+            # Clear state if original image load fails
+            st.session_state.enhanced_image_data = None
+            st.session_state.error_message = None
+            st.session_state.current_file_identifier = None
+            enhance_button_pressed = False # Ensure button state is false if image fails
+            st.stop() # Stop execution if image can't be loaded
+
+    # Process image button and display enhanced result area
+    with col2:
+        st.subheader("Enhanced Result")
+
+        # Use a container within col2 for dynamic content (spinner, result)
+        # Button is now outside this container, in col1
+        result_container = st.container()
+
+        with result_container:
+            # Trigger processing logic if button in col1 was pressed
+            if enhance_button_pressed:
+                st.session_state.enhanced_image_data = None # Clear previous result before trying again
+                st.session_state.error_message = None      # Clear previous errors
+                # Show spinner *within the container*
+                with st.spinner("Enhancing your image... This might take a moment."):
+                    try:
+                        # Prepare form data for the API request
+                        # Use getvalue() to read file content for the request
+                        files = {"file": (uploaded_file.name, uploaded_file.getvalue(), uploaded_file.type)}
+                        payload = {
+                            "model_name": selected_model,
+                            "outscale": float(output_scale),
+                            "face_enhance": face_enhance,
+                            "fp32": use_fp32
+                        }
+
+                        # Send request to API
+                        start_time = time.time()
+                        response = requests.post(f"{API_URL}/enhance/", files=files, data=payload, timeout=300)
+                        end_time = time.time()
+
+                        if response.status_code == 200:
+                            # --- Store results in session state on success ---
+                            st.session_state.enhanced_image_data = response.content
+                            st.session_state.enhanced_image_caption = f"Enhanced ({end_time - start_time:.2f}s)"
+
+                            # Prepare filename for download
+                            base, ext = Path(uploaded_file.name).stem, Path(uploaded_file.name).suffix
+                            download_filename = f"{base}_enhanced_s{int(output_scale)}x{ext if ext else '.png'}" # Use int scale
+                            st.session_state.download_filename = download_filename
+                            st.session_state.download_mime = response.headers.get("content-type", "image/png")
+                            st.session_state.error_message = None # Clear error on success
+
+                        else:
+                            # Store error details from JSON response
+                            try:
+                                error_details = response.json().get('detail', 'Unknown API error')
+                            except requests.exceptions.JSONDecodeError:
+                                error_details = response.text # Fallback to raw text
+                            st.session_state.error_message = f"API Error (Status {response.status_code}): {error_details}"
+                            st.session_state.enhanced_image_data = None # Ensure data is cleared on error
+
+                    except requests.exceptions.Timeout:
+                         st.session_state.error_message = "Error: The enhancement request timed out. The process may be too long or the server might be overloaded."
+                         st.session_state.enhanced_image_data = None
+                    except requests.exceptions.RequestException as e:
+                         st.session_state.error_message = f"Error connecting to API: {e}"
+                         st.session_state.enhanced_image_data = None
+                    except Exception as e:
+                         st.session_state.error_message = f"An unexpected error occurred: {e}"
+                         st.session_state.enhanced_image_data = None
+                    # Spinner automatically stops here
+
+            # --- Display results/errors/placeholder also within the container ---
+            # Display Error Message (if any)
+            if st.session_state.error_message:
+                st.error(st.session_state.error_message)
+
+            # Display Enhanced Image and Download Button (if available in state)
+            # Use elif to prevent showing placeholder if error exists or image exists
+            elif st.session_state.enhanced_image_data is not None:
+                try:
+                    enhanced_image = Image.open(io.BytesIO(st.session_state.enhanced_image_data))
+                    st.image(
+                        enhanced_image,
+                        use_column_width=True,
+                        caption=st.session_state.enhanced_image_caption
+                    )
+                    st.download_button(
+                        label="Download Enhanced Image 💾",
+                        data=st.session_state.enhanced_image_data,
+                        file_name=st.session_state.download_filename,
+                        mime=st.session_state.download_mime
+                    )
+                except Exception as e:
+                    st.error(f"Error displaying enhanced image: {e}")
+                    # Clear state if display fails to prevent repeated errors
+                    st.session_state.enhanced_image_data = None
+                    st.session_state.error_message = "Failed to display the enhanced image."
+            
+            # Display placeholder only if button wasn't pressed in this run AND no image/error
+            # Note: enhance_button_pressed is defined in col1 now, but its state persists across the rerun
+            elif not st.session_state.get('enhanced_image_data') and not st.session_state.get('error_message'):
+                st.markdown("Click 'Enhance Image ✨' (below Original) to process.")
+
+else:
+    st.info("Upload an image to begin the enhancement process.")
+    # --- Clear state if no file is uploaded ---
+    st.session_state.enhanced_image_data = None
+    st.session_state.enhanced_image_caption = None
+    st.session_state.download_filename = None
+    st.session_state.download_mime = None
+    st.session_state.current_file_identifier = None
+    st.session_state.error_message = None
+
+# --- Footer ---
+st.markdown("---")
+st.markdown("Powered by Real-ESRGAN") # Removed link as requested 

environment.yml

@@ -0,0 +1,23 @@
+name: esrgan-env
+channels:
+  - defaults
+  - conda-forge
+dependencies:
+- pip
+- python=3.10
+- pytorch::pytorch=1.11.0
+- pytorch::torchvision
+- pip:
+  - opencv-python==4.11.0.86
+  - PyYAML
+  - tqdm
+  - yapf
+  - basicsr-fixed
+  - facexlib
+  - gfpgan
+
+  - fastapi==0.104.0
+  - uvicorn==0.23.2
+  - streamlit==1.27.0
+  - python-multipart==0.0.6
+  - psutil

run.py

@@ -0,0 +1,352 @@
+import os
+import subprocess
+import sys
+import time
+import webbrowser
+import requests
+import signal
+import psutil
+from pathlib import Path
+import platform # Import platform module
+
+
+def main():
+    """Run the Image Enhancer Application by starting API and Streamlit app."""
+    # Change to the directory of this script
+    script_dir = Path(__file__).parent.resolve()
+    os.chdir(script_dir)
+
+    print("\n🚀 Starting Image Enhancer Application...\n")
+
+    # Define URLs and Ports
+    api_port = 8000
+    streamlit_port = 8501
+    api_url = f"http://localhost:{api_port}"
+    streamlit_url = f"http://localhost:{streamlit_port}"
+
+    api_process = None
+    streamlit_process = None
+
+    # --- Helper: Kill Process Tree ---
+    def kill_proc_tree(pid, sig=signal.SIGTERM, include_parent=True, timeout=None, on_terminate=None):
+        """Kill a process tree (including grandchildren) with signal `sig` and fallback signal `signal.SIGKILL`.
+           Source: https://psutil.readthedocs.io/en/latest/#kill-process-tree
+        """
+        assert pid != os.getpid(), "won't kill myself"
+        try:
+            parent = psutil.Process(pid)
+            children = parent.children(recursive=True)
+            if include_parent:
+                children.append(parent)
+            for p in children:
+                try:
+                    p.send_signal(sig)
+                except psutil.NoSuchProcess:
+                    pass
+            gone, alive = psutil.wait_procs(children, timeout=timeout, callback=on_terminate)
+            if alive:
+                # Fallback to SIGKILL for processes that didn't terminate
+                for p in alive:
+                    try:
+                        p.kill()
+                    except psutil.NoSuchProcess:
+                        pass
+                psutil.wait_procs(alive, timeout=1) # Wait a bit more
+        except psutil.NoSuchProcess:
+             pass # Process already gone
+
+    # --- Helper: Clean Up Existing Processes ---
+    def cleanup_existing_processes():
+        print("🧹 Checking for and cleaning up existing related processes...")
+        killed_count = 0
+        current_pid = os.getpid()
+        for proc in psutil.process_iter(['pid', 'name', 'cmdline']):
+            if proc.info['pid'] == current_pid:
+                continue # Don't kill self
+            try:
+                cmdline = proc.info['cmdline']
+                if not cmdline: continue
+
+                # Check for Uvicorn running api.py on the specified port
+                is_api = (
+                    ('uvicorn' in proc.info['name'] or 'python' in proc.info['name']) and
+                    any(f'api:app' in arg for arg in cmdline) and
+                    any(f'--port={api_port}' in arg or f'--port {api_port}' in arg for arg in cmdline)
+                )
+
+                # Check for Streamlit running app.py
+                is_streamlit = (
+                    ('streamlit' in proc.info['name'] or 'python' in proc.info['name']) and
+                    any('streamlit' in arg and 'run' in arg and 'app.py' in arg for arg in cmdline)
+                )
+
+                if is_api or is_streamlit:
+                    service_name = "API server" if is_api else "Streamlit app"
+                    print(f"   Killing existing {service_name} (PID: {proc.info['pid']}) {' '.join(cmdline)[:80]}...")
+                    kill_proc_tree(proc.info['pid']) # Kill the process and its children
+                    killed_count += 1
+
+            except (psutil.NoSuchProcess, psutil.AccessDenied, psutil.ZombieProcess):
+                continue # Process might have died already
+            except Exception as e:
+                 print(f"   Error checking process {proc.info['pid']}: {e}")
+        if killed_count > 0:
+             print(f"   Killed {killed_count} existing process(es).")
+        else:
+            print("   No conflicting processes found.")
+        time.sleep(1) # Give OS a moment to release ports
+
+    # --- Start API Server ---
+    def start_api_server():
+        nonlocal api_process
+        print(f"🔄 Starting API server (api.py) on port {api_port}...")
+        cmd = [sys.executable, "api.py"] # Assuming api.py handles its own uvicorn run
+        try:
+            # Use Popen for non-blocking execution
+            api_process = subprocess.Popen(
+                cmd,
+                stdout=subprocess.PIPE,
+                stderr=subprocess.PIPE,
+                bufsize=1, # Line buffered
+                universal_newlines=True,
+                creationflags=subprocess.CREATE_NEW_PROCESS_GROUP if platform.system() == "Windows" else 0 # Create new group for easier cleanup on Windows
+            )
+            print(f"   API server process started (PID: {api_process.pid})")
+        except Exception as e:
+            print(f"❌ Failed to start API server: {e}")
+            api_process = None
+        return api_process
+
+    # --- Check API Health ---
+    def check_api_health(timeout=3):
+        if not api_process or api_process.poll() is not None:
+             return False # Process not running
+        try:
+            response = requests.get(f"{api_url}/status", timeout=timeout)
+            return response.status_code == 200 and response.json().get("status") in ["ok", "busy"]
+        except requests.exceptions.RequestException:
+            return False
+
+    # --- Start Streamlit App ---
+    def start_streamlit_app():
+        nonlocal streamlit_process
+        print(f"🔄 Starting Streamlit web interface (app.py) on port {streamlit_port}...")
+        cmd = [sys.executable, "-m", "streamlit", "run", "app.py", f"--server.port={streamlit_port}", "--server.headless=true"]
+        try:
+            streamlit_process = subprocess.Popen(
+                cmd,
+                stdout=subprocess.PIPE,
+                stderr=subprocess.PIPE,
+                bufsize=1,
+                universal_newlines=True,
+                creationflags=subprocess.CREATE_NEW_PROCESS_GROUP if platform.system() == "Windows" else 0
+            )
+            print(f"   Streamlit process started (PID: {streamlit_process.pid})")
+        except Exception as e:
+            print(f"❌ Failed to start Streamlit app: {e}")
+            streamlit_process = None
+        return streamlit_process
+
+    # --- Shutdown Services ---
+    def shutdown_services(api_proc, streamlit_proc):
+        print("\n🛑 Shutting down services...")
+        processes_to_stop = {
+            "API": api_proc,
+            "Streamlit": streamlit_proc
+        }
+        for name, proc in processes_to_stop.items():
+            if proc and proc.poll() is None:
+                print(f"   Stopping {name} (PID: {proc.pid})...")
+                try:
+                    # Use the kill_proc_tree helper for robust termination
+                    kill_proc_tree(proc.pid, sig=signal.SIGTERM, timeout=3)
+                    print(f"   {name} stopped.")
+                except Exception as e:
+                    print(f"   Error stopping {name} (PID: {proc.pid}): {e}. Attempting force kill.")
+                    try:
+                       kill_proc_tree(proc.pid, sig=signal.SIGKILL)
+                    except Exception as final_e:
+                         print(f"   Force kill also failed for {name} (PID: {proc.pid}): {final_e}")
+            elif proc:
+                print(f"   {name} (PID: {proc.pid}) already stopped.")
+            else:
+                 print(f"   {name} was not running.")
+        # Final cleanup check
+        cleanup_existing_processes()
+        print("✅ Application stopped.")
+
+    # --- Main Execution Logic ---
+
+    # Set up graceful exit handler
+    def handle_exit(signum, frame):
+        print("\n👋 Signal received, initiating graceful shutdown...")
+        shutdown_services(api_process, streamlit_process)
+        sys.exit(0)
+
+    signal.signal(signal.SIGINT, handle_exit) # Ctrl+C
+    signal.signal(signal.SIGTERM, handle_exit) # Termination signal
+
+    try:
+        # Initial cleanup before starting
+        cleanup_existing_processes()
+
+        # Start API
+        api_process = start_api_server()
+        if not api_process:
+             raise RuntimeError("Failed to start API server, cannot continue.")
+
+        # Wait for API to be ready
+        print("   Waiting for API to become available", end="")
+        api_ready = False
+        for i in range(20): # Increased wait time (20 secs)
+            if check_api_health():
+                api_ready = True
+                print("\n✅ API server is running and responding.")
+                break
+            print(".", end="", flush=True)
+            time.sleep(1)
+            if api_process.poll() is not None:
+                print(f"\n❌ API process terminated unexpectedly during startup (exit code: {api_process.poll()}).")
+                # Attempt to read stderr
+                try:
+                     _, stderr_output = api_process.communicate(timeout=1)
+                     print("--- API Stderr ---")
+                     print(stderr_output or "<No stderr captured>")
+                     print("------------------")
+                except:
+                     pass
+                raise RuntimeError("API process failed during startup.")
+
+        if not api_ready:
+            print("\n⚠️ API server did not become responsive within the time limit. Check api.log. Proceeding anyway...")
+
+        # Start Streamlit
+        streamlit_process = start_streamlit_app()
+        if not streamlit_process:
+            print("❌ Failed to start Streamlit app. You may need to start it manually.")
+            # Don't raise error, maybe user only wants API
+
+        # Give Streamlit a moment
+        print("   Waiting for Streamlit to initialize...")
+        time.sleep(5)
+        if streamlit_process and streamlit_process.poll() is None:
+            print("✅ Streamlit interface should be starting.")
+        elif streamlit_process:
+            print(f"❌ Streamlit process terminated unexpectedly after start (exit code: {streamlit_process.poll()}). Check logs.")
+        else:
+            print("   Streamlit process failed to start.")
+
+        # Open browser if Streamlit started
+        if streamlit_process and streamlit_process.poll() is None:
+            try:
+                print(f"   Opening web interface ({streamlit_url}) in your browser...")
+                webbrowser.open(streamlit_url)
+            except Exception as e:
+                print(f"   Could not open browser automatically: {e}. Please navigate to the URL manually.")
+        else:
+             print(f"   Streamlit not running, cannot open browser.")
+
+        # Print URLs
+        print("\n📋 Application URLs:")
+        print(f"   - Web Interface: {streamlit_url} (if started)")
+        print(f"   - API Root: {api_url}")
+        print(f"   - API Status: {api_url}/status")
+        print(f"   - API Models: {api_url}/models")
+
+        print("\n✨ Application is running.")
+        print("   Monitoring services... Press Ctrl+C to stop.")
+        print("   Check api.log for API server logs.")
+
+        # Monitoring Loop
+        last_health_check_time = time.time()
+        consecutive_api_failures = 0
+        while True:
+            # Check API process
+            api_status = api_process.poll() if api_process else -1
+            if api_status is not None:
+                print(f"\n⚠️ API server process stopped unexpectedly (exit code: {api_status}). Restarting...")
+                # Attempt to read stderr
+                try:
+                     _, stderr_output = api_process.communicate(timeout=1)
+                     print("--- API Stderr ---")
+                     print(stderr_output or "<No stderr captured>")
+                     print("------------------")
+                except:
+                     pass
+                api_process = start_api_server()
+                if not api_process:
+                     print("❌ Failed to restart API server after crash. Exiting monitoring loop.")
+                     break
+                time.sleep(5) # Give it time to restart
+                last_health_check_time = time.time() # Reset check timer
+                consecutive_api_failures = 0
+                continue # Skip rest of loop iteration
+
+            # Check Streamlit process
+            if streamlit_process:
+                streamlit_status = streamlit_process.poll()
+                if streamlit_status is not None:
+                    print(f"\n⚠️ Streamlit process stopped unexpectedly (exit code: {streamlit_status}). Restarting...")
+                     # Attempt to read stderr
+                    try:
+                         _, stderr_output = streamlit_process.communicate(timeout=1)
+                         print("--- Streamlit Stderr ---")
+                         print(stderr_output or "<No stderr captured>")
+                         print("----------------------")
+                    except:
+                          pass
+                    streamlit_process = start_streamlit_app()
+                    if not streamlit_process:
+                        print("❌ Failed to restart Streamlit. Will not monitor Streamlit anymore.")
+                    time.sleep(5) # Give it time to restart
+                    continue
+
+            # Periodic Health Check (every 30 seconds)
+            current_time = time.time()
+            if current_time - last_health_check_time > 30:
+                if check_api_health():
+                    # print("   [Health Check] API is responsive.") # Verbose logging
+                    consecutive_api_failures = 0
+                else:
+                    consecutive_api_failures += 1
+                    print(f"   [Health Check] ⚠️ API failed health check #{consecutive_api_failures}.")
+                    if consecutive_api_failures >= 3:
+                        print("   API unresponsive for 3 consecutive checks. Restarting API server...")
+                        kill_proc_tree(api_process.pid) # Force kill unresponsive API
+                        api_process = start_api_server()
+                        if not api_process:
+                            print("❌ Failed to restart API server after health failures. Exiting monitoring loop.")
+                            break
+                        time.sleep(5)
+                        consecutive_api_failures = 0 # Reset counter after restart
+
+                last_health_check_time = current_time
+
+            # Add memory check if needed (optional)
+            # try:
+            #    api_mem = psutil.Process(api_process.pid).memory_info().rss / (1024 * 1024) # MB
+            #    if api_mem > 1024: # Example threshold: 1GB
+            #        print(f"   ⚠️ API memory usage high ({api_mem:.1f} MB). Restarting...")
+            #        kill_proc_tree(api_process.pid)
+            #        api_process = start_api_server()
+            #        # ... handle failure ...
+            # except psutil.NoSuchProcess:
+            #    pass # Process might have just restarted
+
+            time.sleep(5) # Check processes every 5 seconds
+
+    except KeyboardInterrupt:
+        print("\n⌨️ Ctrl+C detected.")
+    except RuntimeError as e:
+         print(f"\n❌ Runtime Error: {e}")
+    except Exception as e:
+        print(f"\n❌ An unexpected error occurred in the run script: {e}")
+        import traceback
+        traceback.print_exc()
+    finally:
+        # Ensure services are stopped on any exit
+        shutdown_services(api_process, streamlit_process)
+
+
+if __name__ == "__main__":
+    main()