Integrating AI models into Unreal Engine

This thesis explores the integration of artificial intelligence (AI) models into virtual reality (VR) and augmented reality (AR) applications, with a focus on Meta Quest headsets. The project aims to overcome the limitations of Meta's native APIs, which restrict access to the headset's camera feed, and instead develops an external camera solution for real-time AI processing. The system captures real-world visual data through an external camera, processes it using AI models on a dedicated server, and integrates the results into the game environment in Unreal Engine. By utilizing WebSocket connections for real-time data transmission, this approach enhances the interactivity of VR/AR applications by enabling real-time AI-driven updates. The results demonstrate a significant improvement in the responsiveness and immersion of VR/AR environments, opening new possibilities for gaming and non-gaming applications, such as real-time data analysis and monitoring. Future work includes advancing emotional analysis models for AR/VR users and improving camera synchronization between external and headset cameras.