🎨 Rembg Gradio Interface

A clean Gradio interface for Rembg, inspired by the original Rembg server implementation but designed as a standalone, Docker-friendly application.

✨ Features

• 🎯 Multiple AI Models: Choose from 15+ specialized models for different use cases
• 🖼️ Live Preview: See results instantly with side-by-side comparison
• ⚙️ Advanced Options: Fine-tune with alpha matting and post-processing controls
• 🐳 Docker Ready: Easy deployment with included Dockerfile and docker-compose
• 🚀 GPU Acceleration: Optimized for CUDA-enabled GPUs
• 📁 Batch Processing: Process multiple images efficiently

🚀 Quick Start

Local Installation

1. Clone this repository: bash git clone https://github.com/yourusername/rembg.git cd rembg

2. Install dependencies: bash pip install rembg[gpu] gradio

3. Download the default model: bash rembg d u2net

4. Launch the application: bash python app.py

5. Open your browser and navigate to http://localhost:7860

🐳 Docker Deployment

Build and run with Docker Compose:

bash docker-compose up --build

The interface will be available at http://localhost:7860

🎮 Usage Guide

Basic Usage

1. Upload Image: Drag and drop or click to upload an image
2. Select Model: Choose from available models (default: u2net)
3. Remove Background: Click the button to process
4. Download Result: Save the output image with transparent background

Available Models

U2Net Family

• 🎯 u2net (default): A pre-trained model for general use cases
• ⚡ u2netp: A lightweight version of u2net model
• 👤 u2nethumanseg: A pre-trained model for human segmentation
• 👔 u2netclothseg: A pre-trained model for clothes parsing from human portrait

ISNet Models

• 🔍 isnet-general-use: A new pre-trained model for general use cases
• 🎌 isnet-anime: A high-accuracy segmentation for anime character

BiRefNet Family

• ✨ birefnet-general: A pre-trained model for general use cases
• 🚀 birefnet-general-lite: A light pre-trained model for general use cases
• 👨 birefnet-portrait: A pre-trained model for human portraits
• 🎯 birefnet-dis: A pre-trained model for dichotomous image segmentation (DIS)
• 🔬 birefnet-hrsod: A pre-trained model for high-resolution salient object detection (HRSOD)
• 🕵️ birefnet-cod: A pre-trained model for concealed object detection (COD)
• 💪 birefnet-massive: A pre-trained model with massive dataset

Other Models

• 🤖 sam: A pre-trained model for any use cases
• 🎨 silueta: Same as u2net but the size is reduced to 43Mb
• 🏢 bria-rmbg: Commercial-grade background removal model (Bria AI)

Advanced Options

• Alpha Matting: Improves edge quality for hair and fur

  • Foreground Threshold: Higher values keep more foreground
  • Background Threshold: Lower values remove more background
  • Erode Size: Shrinks the foreground mask

• Output Options:

  • Mask Only: Output segmentation mask instead of transparent image
  • Post Process: Apply morphological operations to clean the mask

⚙️ Configuration

Environment Variables

Create a .env file for docker-compose:

env PUBLIC_PORT=7860:7860 REPLICAS_COUNT=1

GPU Support

For GPU acceleration, ensure you have: - NVIDIA GPU with CUDA support - nvidia-docker installed - --gpus all flag when running Docker

🔧 Development

Project Structure

rembg/ ├── app.py # Gradio interface ├── Dockerfile # Docker configuration ├── docker-compose.yml # Docker Compose setup ├── requirements.txt # Python dependencies └── examples/ # Sample images

Customization

To add new features or modify the interface, edit app.py. The main components are:

• process_image(): Core processing function
• gr.Blocks(): Gradio interface layout
• Model selection and parameter controls

📊 Performance Tips

• 🚀 Model Selection: Start with u2net for general images, use specialized models for specific content
• ⚡ GPU Usage: Ensure CUDA is available for 10x+ speedup
• 🎯 Alpha Matting: Enable only when needed (adds processing time)
• 📦 Batch Processing: Process multiple similar images with the same model loaded

🙏 Acknowledgements

• Rembg by Daniel Gatis for the core background removal functionality
• Gradio for the web interface framework
• All the model authors whose work powers the background removal

📄 License

This project is licensed under the MIT License - see the original Rembg LICENSE for details.

Made with ❤️ inspired by the original Rembg project