RT-DETR TensorRT Engine Deployment

This project demonstrates how to deploy the RT-DETR model using C++ and TensorRT. For the model details, please refer to the RT-DETR GitHub Repository.

Prerequisites

Before getting started, ensure that the following software versions are installed:

• TensorRT: 10.0.1.6
• OpenCV: 4.8.1
• CUDA: 12.3

Project Overview

RT-DETR is a high-performance real-time detection model that is deployed using TensorRT to maximize inference efficiency. This project provides a clean and optimized C++ implementation to help users integrate RT-DETR into their own applications.

Setup

To get started with this project, make sure the required dependencies are installed, and follow the steps below to build and run the deployment engine.

1. Clone the repository:

   git clone https://github.com/Anunnaki-Panda/RTDETR-TensorRT-Engine
   cd RTDETR-TensorRT-Engine
   

2. Build the project with CMake:

   mkdir build && cd build
   cmake ..
   make

3. Run the deployment:

   ./RTDETR