Car Background Removal API

A FastAPI application that detects vehicles in images, crops them, and removes the background using AI models.

Features

• Vehicle detection using YOLO11
• Automatic cropping to vehicle bounds
• Background removal using rembg
• RESTful API endpoint

Installation

1. Clone or download this project

2. Install Python dependencies

   pip install -r requirements.txt

3. Run the application

   uvicorn app:app --reload

   Or with specific host and port:

   uvicorn app:app --host 0.0.0.0 --port 8000 --reload

Usage

Once the server is running, you can:

1. Visit the interactive API docs: http://localhost:8000/docs
2. Make POST requests to /car-remove-background/ with an image file

Example using curl:

curl -X POST "http://localhost:8000/car-remove-background/" \
     -H "accept: application/json" \
     -H "Content-Type: multipart/form-data" \
     -F "file=@your_car_image.jpg" \
     --output result.png

Example using Python requests:

import requests

url = "http://localhost:8000/car-remove-background/"
with open("car_image.jpg", "rb") as f:
    files = {"file": f}
    response = requests.post(url, files=files)

with open("result.png", "wb") as f:
    f.write(response.content)

Docker

Build a CPU-only image and run the API on port 8000.

# Build
docker build -t car-bg-api:latest .

# Run
docker run --rm -p 8000:8000 \
  -e OMP_NUM_THREADS=8 \
  car-bg-api:latest

Notes:

• If yolo11x.pt exists in the project root, it will be included in the image and used directly.
• If the weight is missing, Ultralytics will auto-download it at first request.
• On low-memory hosts, you can lower the number of threads via OMP_NUM_THREADS.

Notes

• The YOLO model will be automatically downloaded on first run (if not bundled)
• Supported image formats: JPG, PNG, etc.
• The API returns a PNG image with transparent background
• Vehicles (cars, motorcycles, buses, trucks) are detected and processed