Dockerized Background Removal Web App - Quick Start Guide#
Run privacy-focused AI background removal locally with Docker. One-command setup, offline processing, no API key needed. Perfect for product photos, profile pictures, and secure image editing.
What It Is#
A simple web app that removes backgrounds from your photos using AI.
Just upload a photo, and get back a clean image with the background removed. Perfect for product photos, profile pictures, or any image where you want to remove the background.
What You Need#
Just Docker installed on your computer. That's it!
System Requirements#
- Docker: Docker Desktop (Windows/Mac) or Docker Engine (Linux)
- RAM: Minimum 4GB (8GB recommended for faster processing)
- Storage: 2GB free disk space for Docker image
- OS: Windows 10+, macOS 10.15+, or modern Linux distribution
How to Start#
Install Docker, run this command, then open http://localhost. Your photos never leave your machine.
docker run -p 80:80 withoutbg/app:latestThis command downloads the Docker image (first run only, ~2GB) and starts the web app on port 80.
Open your browser and navigate to http://localhost
How to Use#
- Drag and drop a photo into the upload area (or click to browse)
- Wait a few seconds while the AI removes the background
- Download your result!
Supported Formats#
The app works with:
- Image formats: JPEG, PNG, WebP
- File size: Up to 10MB per image
- Processing: One image at a time (typically 2-5 seconds)
- Output: Transparent PNG with preserved quality
What's Inside#
Behind the scenes, this web app contains:
- Modern web interface: Drag-and-drop upload with real-time processing feedback
- AI models: 4-stage pipeline (Depth → ISNet → Matting → Refiner) using open-source Focus models (~320MB)
- Privacy-first processing: All computation happens in RAM, files discarded immediately after processing
- Cross-platform: Containerized with Docker for consistent behavior on Windows, Mac, and Linux
Everything is pre-configured and ready to use. No API keys, no configuration files, no setup required.
Performance#
Typical performance metrics on modern hardware:
- First startup: 30-60 seconds (one-time Docker image download, ~2GB)
- Container initialization: 10-15 seconds (AI models load into memory)
- Image processing: 2-5 seconds per image (CPU dependent)
- Memory usage: ~2-3GB RAM during processing
Troubleshooting#
Common Issues#
Port already in use
If port 80 is already in use, you can map to a different port:
Alternative Port
docker run -p 8080:80 withoutbg/app:latestThen access at http://localhost:8080
Docker daemon not running
Make sure Docker Desktop is running. Look for the Docker icon in your system tray (Windows) or menu bar (Mac).
Cannot connect to localhost
Wait 10-15 seconds after running the command for the container to fully start. Check container status:
Check Container Status
docker psSlow processing
Ensure Docker has sufficient resources allocated. In Docker Desktop, go to Settings → Resources and allocate at least 4GB RAM and 2 CPU cores.
Advanced Docker Commands#
Advanced Commands
# Run in detached mode (background)
docker run -d -p 80:80 withoutbg/app:latest
# Stop the container
docker stop $(docker ps -q --filter ancestor=withoutbg/app:latest)
# View logs
docker logs $(docker ps -q --filter ancestor=withoutbg/app:latest)
# Remove container and image
docker rm -f $(docker ps -aq --filter ancestor=withoutbg/app:latest)
docker rmi withoutbg/app:latestFrequently Asked Questions#
Is my data private when using the Docker web app?#
Yes. Everything runs on your local machine. Your photos never leave your computer, are processed in RAM, and discarded immediately after processing. No cloud storage, no external API calls, no data collection.
What are the system requirements?#
You need Docker Desktop (for Windows/Mac) or Docker Engine (for Linux), at least 4GB RAM, and 2GB free disk space for the Docker image. The app works on Windows 10+, macOS 10.15+, or modern Linux distributions.
Can I process images offline?#
Yes. Once the Docker image is downloaded (first run only), the web app works completely offline. No internet connection or API key required. Perfect for air-gapped environments or when traveling.
How fast is the processing?#
Processing typically takes 2-5 seconds per image on modern hardware, depending on image size and your CPU. First run may take longer as the AI models initialize (one-time setup).
Can I use this for commercial projects?#
Yes. The Docker web app uses open-source models and is free for both personal and commercial use. Check the license terms on the Docker Hub page for details.
Related Resources#
For more advanced integration options:
- Python SDK - Integrate background removal into Python applications with local or cloud processing
- Background Removal API - Direct API access for custom integrations and highest quality results
- Open Source Model Results - See example outputs from the local model used in this Docker app
- Full Documentation - Complete API reference and guides