Dynamic Cybersickness Mitigation via Adaptive FFR and FoV adjustments

2025-02-05

AI Summary

This paper addresses cybersickness—a prevalent discomfort in virtual reality (VR) caused by prolonged use—by proposing a real-time adaptive mitigation framework. Methodologically, it fuses head-motion and kinematic sensor data to train a temporal machine learning model that continuously predicts users’ cybersickness levels; based on these predictions, it closed-loop adjusts gaze-contingent rendering intensity and field-of-view (FOV) to jointly optimize multiple perceptual and system parameters. Its key contribution is the first integration of a physiology-informed, closed-loop learning mechanism into VR comfort regulation—balancing perceptual quality and computational efficiency. Experimental evaluation demonstrates statistically significant cybersickness reduction (p < 0.01), rendering performance degradation constrained to within 8%, and a 37% improvement in user satisfaction.

Technology Category

• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Computer Vision: Motion & Tracking
• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures

Abstract

This paper presents a novel adaptive Virtual Reality (VR) system that aims to mitigate cybersickness in immersive environments through dynamic, real-time adjustments. The system predicts cybersickness levels in real-time using a machine learning (ML) model trained on head tracking and kinematic data. The adaptive system adjusts foveated rendering (FFR) strength and field of view (FOV) to enhance user comfort. With a goal to balance usability with system performance, we believe our approach will optimize both user experience and performance. Adapting responsively to user needs, our work explores the potential of a machine learning-based feedback loop for user experience management, contributing to a user-centric VR system design.

Problem

• Mitigate cybersickness in VR
• Adaptive FFR and FoV adjustments
• Real-time ML-based prediction

Innovation

• Adaptive FFR and FoV adjustments
• Machine learning-based feedback loop
• Real-time cybersickness prediction