README.md

TensorRT-Vision

A TensorRT Toolbox for Optimized Vision Model Inference

📋 Overview

TensorRT-Vision provides optimized inference for computer vision models using NVIDIA TensorRT. It supports:

• Object Detection
• Object Segmentation
• Object Classification
• Object Re-Identification
• Multi Object Tracking

⚙️ Requirements

1. CUDA 12.6
2. TensorRT 10.7.0
3. Python 3.12.3

Follow installation instructions here

🛠️ Build & Install

# Build all apps (default)
meson setup build
meson compile -C build

# Or explicitly specify apps to build
meson setup build -Dbuild_apps=detector,mot
meson compile -C build

# Make sure trtexec is installed for model export
alias trtexec='/usr/src/tensorrt/bin/trtexec'

🚀 Quick Start

Each app has its own README with detailed instructions:

• Object Detection Guide
• Object Segmentation Guide
• Multi Object Tracking Guide
• Object Classification Guide
• Object Re-Identification Guide

🙏 Credits

This project builds upon foundations from:

• tensorrt-cpp-api - A C++ TensorRT wrapper