yolov11-trt

This repository contains a C++ implementation of YOLOv11, optimized with TensorRT for fast and efficient real-time inference.

Dependencies

• C++:

  • OpenCV
  • TensorRT

• Python:

  pip install --upgrade ultralytics

Installation

Repository clone & build

git clone https://github.com/fbasatemur/yolov11-trt.git
cd yolov11-trt*

mkdir build && cd build
cmake ..
cmake --build . --config Release

Usage

Exporting the Model

Modify the export.py script for the desired model name, then run it to export the YOLOv11 model to ONNX.

python export.py

Create a TRT Engine

Convert the ONNX model to a TRT engine:

./yolov11-trt yolo11.onnx "yolov11.engine"

Running Inference

Perform object detection on an image:

./yolov11-trt yolo11.engine "img1.jpg"

Perform object detection on a video:

./yolov11-trt yolo11.engine "vid1.mp4"

References

• https://github.com/spacewalk01/yolov11-tensorrt