suresh-svg-converter 0.2.0

A professional Image to SVG converter with background removal and multi-color support.

---

title: Suresh SVG Converter emoji: 🖼️ colorFrom: blue colorTo: indigo sdk: docker app_port: 7860

🖼️ Suresh SVG Converter

A professional, production-ready Python package for removing backgrounds from images and converting them to high-quality multi-color SVG format.

✨ Features

• Background Removal: AI-powered background removal using rembg with ONNX runtime.
• Accurate SVG Conversion: High-quality vector tracing using potrace with multi-color support.
• Color Quantization: Automatically detects dominant colors and creates stratified SVG layers.
• Dual Interface: Full-featured Command Line Interface (CLI) and a beautiful FastAPI web UI.
• CPU Optimized: Runs entirely on CPU with no GPU dependency.
• Fast Performance: Optimized for speed with lazy model loading and K-Means quantization.
• Batch Processing: Process multiple images at once via CLI.

📋 Requirements

• Python 3.9+
• 2GB+ RAM recommended for high-resolution processing.

🚀 Installation

Install the package directly from PyPI:

pip install suresh-svg-converter

🖥️ Usage

Command Line Interface (CLI)

Once installed, use the suresh-svg-converter command anywhere:

Basic usage:

suresh-svg-converter input.jpg

With options:

# Convert with 16 colors and remove background
suresh-svg-converter input.png --remove-bg --colors 16

# Custom output directory
suresh-svg-converter *.png -o ./results

# Batch process multiple files
suresh-svg-converter img1.jpg img2.webp img3.png

CLI Options:

• --remove-bg, -r: Remove background (default: enabled)
• --colors, -c: Number of colors for SVG tracing (default: 8)
• --output-dir, -o: Custom output directory
• --max-dim: Maximum image dimension for processing (default: 4096)
• --verbose, -v: Enable verbose logging

Web Interface

Launch the interactive web UI to drag and drop images:

python -m suresh_svg_converter.main

Or use uvicorn directly:

# Locally
uvicorn suresh_svg_converter.main:app --host 0.0.0.0 --port 8000

# On Hugging Face (Port 7860)
uvicorn suresh_svg_converter.main:app --host 0.0.0.0 --port 7860

Then open http://localhost:8000 in your browser.

Programmatic API

Use the converter in your own Python projects:

from suresh_svg_converter.services.svg_convert import SVGConverter
from suresh_svg_converter.services.bg_remove import BackgroundRemover
from PIL import Image

# Load image
img = Image.open("photo.jpg")

# Remove background
remover = BackgroundRemover()
img_no_bg = remover.remove_background(img)

# Convert to SVG with 12 colors
converter = SVGConverter()
svg_path = converter.convert(img_no_bg, "output.svg", num_colors=12)

print(f"Vector saved to: {svg_path}")

📁 Project Structure

suresh_svg_converter/
├── cli.py                   # Main CLI entry point
├── main.py                  # FastAPI web application
├── config.py                # Configuration settings
├── services/
│   ├── bg_remove.py         # Background removal service
│   ├── svg_convert.py       # Multi-color SVG conversion
│   └── image_utils.py       # Image preprocessing
├── routes/
│   └── upload.py            # API routes
└── models/
    └── image_models.py      # Data schemas

⚙️ Configuration

You can customize the default behavior in suresh_svg_converter/config.py:

• SVG_DEFAULT_COLORS: Default number of color layers (default: 8)
• MAX_IMAGE_DIMENSION: Resize limit for performance
• REMBG_MODEL: Choose between u2net, u2netp, etc.

📝 License

MIT License - See LICENSE file for details.

🙏 Acknowledgments

• rembg for background removal.
• potrace for vector tracing.
• FastAPI for the web interface.

---

Built with ❤️ for the developer community.

s u r e s h - s v g - c o n v e r t e r