Note: This application is a modified version of the Tile-Upscaler project. Special thanks to the original creator for their work in developing the initial version.
Image Upscale is an AI-powered application designed to enhance and upscale images using advanced techniques like Stable Diffusion and Tile ControlNet. It provides high-quality image enhancement with options for HDR effects and customizable settings.
- Image Upscaling: Increase the resolution of images while maintaining quality.
- Detail Enhancement: Improve image details and overall quality.
- HDR Effects: Apply HDR-like effects for more vibrant and dynamic images.
- Customizable Settings: Adjust resolution, enhancement strength, inference steps, HDR effect, and guidance scale.
- Stable Diffusion with Tile ControlNet: Utilizes advanced AI models for image processing.
- RealESRGAN: Provides high-quality upscaling.
- Custom HDR Processing: Enhances dynamic range for better image quality.
- Python 3.7 or higher
- NVIDIA GPU with CUDA support (recommended for better performance)
- Torch: You will need the following specific versions for GPU support:
torch==2.0.0torchvision==0.15.1torchaudio==2.0.1- Install via:
pip install torch==2.0.0 torchvision==0.15.1 torchaudio==2.0.1 --index-url https://download.pytorch.org/whl/cu118
-
Clone the repository:
git clone https://github.com/SUP3RMASS1VE/Image-Upscale.git cd Image-Upscale -
Install dependencies:
pip install -r requirements.txt
-
Download required models: The application will automatically download the necessary models on the first run.
-
Run the application:
python app.py
-
Access the Gradio interface: Open your web browser and go to http://127.0.0.1:7860.
-
Upload and enhance images:
- Upload an image using the interface.
- Adjust the settings as needed.
- Click "Enhance Image" to process the image.
- FORCE_CPU: Set to
1to force CPU usage even if a GPU is available. - VERBOSE_OUTPUT: Set to
1to enable detailed logging.
- CUDA not available: Ensure that NVIDIA drivers and CUDA-compatible PyTorch are installed correctly.
- Model download issues: Ensure you have a working internet connection and try again if there are download failures.
This project is licensed under the MIT License.
This update ensures that credit is given to the original creator, and the necessary dependency information for GPU support is included. Let me know if you need any further modifications!