ommunicates with Unity via a TCP connection. It uses computer vision techniques to detect facial landmarks, eyes, and mouth features, stabilize their movements, and send this data to Unity for 3D model animation or interaction.