VREmotionDetection

VREmotionDetection is a Unity library for detecting emotions in VR using machine learning models.

📋 Prerequisites

• Unity 6 or newer.
• Meta Quest Pro if you want to use use action units for prediction.

📦 Installation

1. In your Unity project, open Window > Package Manager.

2. In the top-left corner, click + > Install package from git URL...

3. Paste the following URL:

   https://github.com/vcu-swim-lab/VREmotionDetection.git
   

4. Click Install.

⚙️ Configuration

1. In Unity, select the object that you want to add the predictor model from the Hierarchy tab.
2. In the Inspector tab, click Add Component, then Face Au Model.
3. In the Natural Model field, pick natural_trial_96.onnx and for the Acted Model field pick act_trial_92.onnx.

▶ Usage

1. Create a new MonoBehaviour script and attach it to the same GameObject as the Emotion Predictor.
2. Open the script in your code editor.
3. Copy and paste the sample code from:
   Samples~/Basic/MyScript.cs.

This script will give you direct access to the predictions from the emotion detection model.

📝 License

This project is licensed under the MIT License.