ParkingSpace is a research project for real-time parking space detection using YOLOv11 and RTSP cameras. It identifies available parking spaces in urban environments, including areas without pre-defined spots.

Developed under the supervision of Prof. Roi Poranne.

• Real-time detection with YOLOv11
• RTSP camera support
• Works in undefined parking areas
• Probability mapping for space availability

• src/parkingspace/ — Main source code (entry: main.py)
• Demo/ — Demo videos and probability map
• regions.json — Region definitions for parking detection
• requirements.txt / requirements-gpu.txt — Dependencies
• tests/ — Unit tests
• scripts/ — Utility scripts

Requirements:

• Python 3.8+
• 2.5GB free disk space
• CUDA-compatible GPU (optional)

Setup:

# Clone repository
git clone https://github.com/danielbob32/ParkingSpace.git
cd ParkingSpace

# Create and activate virtual environment
python -m venv parkingspace-env
# Windows:
parkingspace-env\Scripts\activate
# Linux/Mac:
source parkingspace-env/bin/activate

# Install dependencies
pip install -r requirements.txt
pip install -e .

# Run demo
python -c "from src.parkingspace import main; main()"

GPU (CUDA 11.7):

pip install -r requirements-gpu.txt
# Or
pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1+cu117 --extra-index-url https://download.pytorch.org/whl/cu117

Check CUDA:

nvcc --version

Run all tests with:

pytest tests/

On first run, models are downloaded automatically. Processing is performed at 15-second intervals by default. Output is shown in a window with detected parking spaces.

• Increase sampling interval for lower resource usage
• Reduce model resolution
• Lower model accuracy for faster processing

Contributions are welcome. Please fork the repository and submit a pull request, or open an issue for suggestions.

1. Fork the repository
2. Create a feature branch
3. Commit changes
4. Push to your fork
5. Open a pull request

Distributed under the Apache License. See LICENSE.

Daniel Bobritski — [email protected]

Project: https://github.com/danielbob32/ParkingSpace

• Home Setup: Home system setup
• How to Record the data: Data acquisition
• Training the system: Custom training
• Effective Regions: Region selection
• Contributing: Contribution guide
• Discord: Community support

• Raw Images: Raw images from live videos
• Augmented Images: Augmented images for probability map
• Segmented Images: YOLOv11 segmented images
• Binary Masks: Binary masks from segmentation
• Probability Map: Probability map from binary masks

• Ultralytics
• Roboflow