SketchCleanUp

A tool for cleaning up and refining sketches using techniques from Mastering Sketch. Provides an interactive Gradio-based UI for users to upload, process, and download cleaned sketches.

Features

Automatic Stroke Refinement: Smooth and normalize sketch strokes for a cleaner look.
Noise Reduction: Remove unwanted artifacts and stray marks.
Adaptive Thresholding: Enhance line clarity based on local pixel intensity.
User Controls: Adjustable parameters for fine-tuning cleanup.
Batch Prediction: leverages GPU availability for Batch generation.
Global Input: the models work well with both pencil sketches and digital sketches.
Interactive UI: Powered by Gradio for easy, browser-based interaction.

Installation

Clone the repository

huggingface-cli login
huggingface-cli download VoidZeroe/XT8 --local-dir SketchCleanUp

Navigate to project directory
```
cd SketchCleanUp
```
Install requirements
```
pip install -r requirements.txt
```

Usage

Launch the Gradio UI by running:

python main.py

Once started, open the provided local URL in your browser. You can then:

Upload your sketch image (supported formats: PNG, JPG, BMP).
Adjust cleanup parameters (e.g., smoothing strength, threshold level).
Click Process to generate the cleaned sketch.
Download the result.

Contributing

Contributions are welcome! Please open issues or submit pull requests for bug fixes, new features, or improvements.

Fork the repository
Create a feature branch (git checkout -b feature/my-feature)
Commit your changes (git commit -m "Add my feature")
Push to the branch (git push origin feature/my-feature)
Open a Pull Request

Acknowledgments

Based research by Mastering Sketch Research
Built with Gradio and OpenCV