Lane Detection for ETS2 using PyTorch LaneNet & YOLOPv2

This project is designed for lane detection in Euro Truck Simulator 2 (ETS2) using deep learning models implemented with PyTorch. It combines two distinct approaches for robust real-time scene understanding:

• Lane Detection using LaneNet (with an ENet backbone) and Object Detection with YOLO11n
• Object Detection and Vehicle Detection using YOLOPv2

YOLOPv2 Preview

Yolopv2.mp4

LaneNet (ENet) + YOLO11n Preview

Lanenet.mp4

Features

• Lane Detection: A PyTorch-based implementation using LaneNet (with an ENet backbone) for accurate lane marking detection, or an alternative approach using YOLOPv2.
• Object Detection: Integration of YOLOPv2 to identify key objects in the driving scene, with an alternative version utilizing YOLO11n.
• Real-Time Processing: Capturing game screen data from ETS2 and processing it in real time.
• Modular Codebase: A clearly separated structure for lane detection, object detection, and utility modules.

Installation

1. Clone the Repository:

   git clone https://github.com/AkramOM606/Lane-Detection.git
   cd Lane-Detection

2. Set Up a Virtual Environment (Optional but Recommended):

   python -m venv venv
   source venv/bin/activate  # For Windows use: venv\Scripts\activate

3. Install Dependencies:

   pip install -r requirements.txt

Usage

Important

Make sure ETS2 is running at 1280x720 in Windowed mode when you execute this script.

Run Lane Detection (LaneNet with ENet):

python main_Lanenet_ENet.py

Run Object Detection (YOLOPv2):

python main_yolopv2.py

Repository Structure

Lane-Detection/
├── lanenet/                # Model definitions and configurations for LaneNet
├── utils/                  # Utility scripts and helper functions
├── weights/                # Pre-trained model weights for both detection methods
├── game_capture.py         # Script to capture game screen data from ETS2
├── lane_detector.py        # Lane detection module
├── main_Lanenet_ENet.py    # Entry point for running LaneNet with ENet
├── main_yolopv2.py         # Entry point for running YOLOPv2-based object detection
├── object_detection.py     # Core object detection functionality
├── yolopv2_detector.py     # YOLOPv2 detection helper
├── requirements.txt        # Python dependencies
└── README.md               # This file

Configuration

Note

1. Pre-trained Weights: Ensure that the required pre-trained model weights are placed in the weights/ directory.
2. Customization: Adjust configuration parameters (e.g., file paths, model hyperparameters) in the scripts as needed for your local setup.

Contributing

Contributions are greatly appreciated! To contribute:

1. Fork this repository.
2. Create a new branch for your feature or bug fix.
3. Commit your changes with clear messages.
4. Submit a pull request for review.

Please adhere to the project's coding standards and update the documentation as needed.

License

This project is licensed under the MIT License.

Acknowledgements

A huge thanks to these awesome projects for providing their incredible models:

• YOLOPv2 by CAIC-AD – for its advanced multi-task perception model.
• LaneNet-Lane-Detection-PyTorch by IrohXu – for the efficient lane detection model.
• And also the ETS2 community for inspiration.

Their contributions have been invaluable in advancing this project!

Disclaimer

This project is a work in progress and is provided "as is" without any warranties. Use at your own risk.