Unity Barracuda Inference YOLOX

This Unity package extends the functionality of the barracuda-inference-base package to perform object detection using YOLOX models. The package contains the YOLOXObjectDetector class, which handles model execution, processes the output, and generates bounding boxes with corresponding labels and colors. It supports various worker types and Asynchronous GPU Readback for model output.

Demo Video

unity-barracuda-inference-yolox-sync-2.mp4

Demo Projects

• barracuda-inference-yolox-demo: A simple Unity project demonstrating how to perform object detection with the barracuda-inference-yolox package using a webcam.
• barracuda-inference-yolox-demo-brp: A simple Unity BRP (Built-in Render Pipeline) project demonstrating how to perform object detection with the barracuda-inference-yolox package using the in-game camera.
• barracuda-inference-yolox-demo-urp: A simple Unity URP project demonstrating how to perform object detection with the barracuda-inference-yolox package using the in-game camera.
• barracuda-inference-yolox-demo-hdrp: A simple Unity HDRP project demonstrating how to perform object detection with the barracuda-inference-yolox package using the in-game camera.

Code Walkthrough

• Code Walkthrough: Unity Barracuda Inference YOLOX Package

Features

• Easy integration with YOLOX object detection models
• Utilizes Unity's Barracuda engine for efficient inference
• Supports various worker types and Asynchronous GPU Readback
• Processes output to generate bounding boxes with labels and colors

Getting Started

Prerequisites

• Unity game engine

Installation

You can install the Barracuda Inference YOLOX package using the Unity Package Manager:

1. Open your Unity project.
2. Go to Window > Package Manager.
3. Click the "+" button in the top left corner, and choose "Add package from git URL..."
4. Enter the GitHub repository URL: https://github.com/cj-mills/unity-barracuda-inference-yolox.git
5. Click "Add". The package will be added to your project.

For Unity versions older than 2021.1, add the Git URL to the manifest.json file in your project's Packages folder as a dependency:

{
  "dependencies": {
    "com.cj-mills.barracuda-inference-yolox": "https://github.com/cj-mills/unity-barracuda-inference-yolox.git",
    // other dependencies...
  }
}

Train a Custom Model

If you're interested in training a YOLOX model on a custom dataset, check out my tutorial tutorial linked below:

• Training YOLOX Models for Real-Time Object Detection in Pytorch

License

This project is licensed under the MIT License. See the LICENSE file for details.