✨ AI Background Remover

Windows:

python -m venv venv
venv\Scripts\activate

Linux/Mac:

python3 -m venv venv
source venv/bin/activate

pip install -r requirements.txt

⏳ First run will download the AI model (~176MB) - this is one-time only!

python main.py

🎉 Success! Open your browser at: http://localhost:8000

import requests

# Remove background from image
with open('photo.jpg', 'rb') as f:
    response = requests.post(
        'http://localhost:8000/remove-background',
        files={'file': f}
    )

# Save result
if response.status_code == 200:
    with open('result.png', 'wb') as output:
        output.write(response.content)
    print("✅ Background removed successfully!")
else:
    print(f"❌ Error: {response.json()['detail']}")

async function removeBackground(imageFile) {
    const formData = new FormData();
    formData.append("file", imageFile);

    const response = await fetch("http://localhost:8000/remove-background", {
        method: "POST",
        body: formData,
    });

    if (response.ok) {
        const blob = await response.blob();
        return URL.createObjectURL(blob);
    }
    throw new Error("Failed to process image");
}

// Usage
const input = document.querySelector('input[type="file"]');
input.addEventListener("change", async (e) => {
    const imageUrl = await removeBackground(e.target.files[0]);
    console.log("✅ Processed image:", imageUrl);
});

curl -X POST "http://localhost:8000/remove-background" \
  -F "file=@photo.jpg" \
  -o result.png

echo "✅ Background removed! Saved as result.png"

Customize the application with environment variables:

# Maximum file size (in bytes)
MAX_FILE_SIZE=10485760  # 10MB

# Port number
PORT=8000

# Host address
HOST=0.0.0.0

Change Port or Host:

# Edit main.py
uvicorn.run("main:app", host="0.0.0.0", port=8000, reload=True)

Change Max File Size:

# Edit main.py
MAX_FILE_SIZE = 10 * 1024 * 1024  # Adjust size

Change Cleanup Schedule:

# Edit main.py
cleanup_old_files(UPLOAD_DIR, max_age_hours=24)  # Adjust hours

Request:

POST /remove-background HTTP/1.1
Content-Type: multipart/form-data

file: [image binary data]

Success Response (200):

HTTP/1.1 200 OK
Content-Type: image/png

[PNG image with transparent background]

Error Responses:

• 400: Invalid file format or corrupted image
• 500: Processing error

Success Response (200):

{
    "status": "healthy",
    "service": "Background Remover API",
    "version": "1.0.0"
}

# Install gunicorn
pip install gunicorn

# Run with multiple workers
gunicorn main:app -w 4 -k uvicorn.workers.UvicornWorker --bind 0.0.0.0:8000

Checklist:

• ✅ Set up reverse proxy (Nginx/Apache)
• ✅ Enable HTTPS (Let's Encrypt)
• ✅ Use process manager (systemd/PM2)
• ✅ Configure firewall rules
• ✅ Set up monitoring

Create Dockerfile:

FROM python:3.10-slim

WORKDIR /app

COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

COPY . .

RUN mkdir -p uploads outputs

EXPOSE 8000

CMD ["python", "main.py"]

Build & Run:

docker build -t bg-remover .
docker run -d -p 8000:8000 --name bg-remover bg-remover

For faster processing with NVIDIA GPU:

# Install ONNX Runtime GPU
pip install onnxruntime-gpu

# Ensure CUDA is installed
nvidia-smi  # Check GPU availability

Performance Boost: ~3-5x faster processing

Use different rembg models:

from rembg import remove, new_session

# Available models
models = ['u2net', 'u2netp', 'u2net_human_seg', 'u2net_cloth_seg', 'silueta']

# Use specific model
session = new_session('u2net_human_seg')
output = remove(input_data, session=session)

Process multiple images:

from pathlib import Path
from rembg import remove

input_dir = Path("input_images")
output_dir = Path("output_images")
output_dir.mkdir(exist_ok=True)

for image_path in input_dir.glob("*.jpg"):
    with open(image_path, 'rb') as f:
        input_data = f.read()

    output_data = remove(input_data)

    output_path = output_dir / f"{image_path.stem}_no_bg.png"
    with open(output_path, 'wb') as f:
        f.write(output_data)

    print(f"✅ Processed: {image_path.name}")