YOLOv11 C++ TensorRT

📜 Citation

The YOLOv11 C++ TensorRT Project is a high-performance object detection solution implemented in C++ and optimized using NVIDIA TensorRT. This project leverages the YOLOv11 model to deliver fast and accurate object detection, utilizing TensorRT to maximize inference efficiency and performance.

📢 Updates

Key Features:

• Model Conversion: Convert ONNX models to TensorRT engine files to accelerate inference.
• Inference on Videos: Efficiently perform object detection on video files.
• Inference on Images: Execute object detection on individual images.
• High Efficiency: Optimized for real-time object detection using NVIDIA GPUs.
• Preprocessing with CUDA: CUDA-enabled preprocessing for faster input handling.

📂 Project Structure

YOLOv11-TensorRT/
├── CMakeLists.txt          # Build configuration for the project
├── include/                # Header files
├── src/
│   ├── main.cpp            # Main entry point for the application
│   ├── yolov11.cpp         # YOLOv11 implementation
│   └── preprocess.cu       # CUDA preprocessing code
├── assets/                 # Images and benchmarks for README
└── build/                  # Compiled binaries

🛠️ Setup

Prerequisites

• CMake (version 3.18 or higher)
• TensorRT (V8.6.1.6: For optimized inference with YOLOv11.)
• CUDA Toolkit (V11.7: For GPU acceleration)
• OpenCV (V4.10.0: For image and video processing)
• NVIDIA GPU (with compute capability 7.5 or higher)

Installation

1. Clone the repository:

   git clone https://github.com/hamdiboukamcha/Yolo-V11-cpp-TensorRT.git
   cd YOLOv11-TensorRT
   

2. Update the TensorRT and OpenCV paths in CMakeLists.txt:

   set(TENSORRT_PATH "F:/Program Files/TensorRT-8.6.1.6")  # Adjust this to your path
   
   

3. Build the project:

   mkdir build
   cd build
   cmake ..
   make -j$(nproc)
   

🚀 Usage

Convert Yolov11 To ONNX Model

from ultralytics import YOLO
Load the YOLO model
model = YOLO("yolo11s.pt")
#Export the model to ONNX format
export_path = model.export(format="onnx")

Convert ONNX Model to TensorRT Engine

To convert an ONNX model to a TensorRT engine file, use the following command:

./YOLOv11TRT convert path_to_your_model.onnx path_to_your_engine.engine.

path_to_your_model.onnx: Path to the ONNX model file.

path_to_your_engine.engine: Path where the TensorRT engine file will be saved.

Run Inference on Video

To run inference on a video, use the following command:

./YOLOv11TRT infer_video path_to_your_video.mp4 path_to_your_engine.engine

path_to_your_video.mp4: Path to the input video file.

path_to_your_engine.engine: Path to the TensorRT engine file.

Run Inference on Video

Run Inference on Image To run inference on an image, use the following command:

./YOLOv11TRT infer_image path_to_your_image.jpg path_to_your_engine.engine

path_to_your_image.jpg: Path to the input image file.

path_to_your_engine.engine: Path to the TensorRT engine file.

⚙️ Configuration

CMake Configuration

In the CMakeLists.txt, update the paths for TensorRT and OpenCV if they are installed in non-default locations:

Set the path to TensorRT installation

#Define the path to TensorRT installation
set(TENSORRT_PATH "F:/Program Files/TensorRT-8.6.1.6")  # Update this to the actual path for TensorRT

Ensure that the path points to the directory where TensorRT is installed.

Troubleshooting

Cannot find nvinfer.lib: Ensure that TensorRT is correctly installed and that nvinfer.lib is in the specified path. Update CMakeLists.txt to include the correct path to TensorRT libraries.

Linker Errors: Verify that all dependencies (OpenCV, CUDA, TensorRT) are correctly installed and that their paths are correctly set in CMakeLists.txt.

Run-time Errors: Ensure that your system has the correct CUDA drivers and that TensorRT runtime libraries are accessible. Add TensorRT’s bin directory to your system PATH.

📞 Contact

For advanced inquiries, feel free to contact me on LinkedIn:

📜 Citation

If you use this code in your research, please cite the repository as follows:

    @misc{boukamcha2024yolov11,
        author = {Hamdi Boukamcha},
        title = {Yolo-V11-cpp-TensorRT},
        year = {2024},
        publisher = {GitHub},
        howpublished = {\url{https://github.com/hamdiboukamcha/Yolo-V11-cpp-TensorRT/}},
    }