Adversarial VR: An Open-Source Testbed for Evaluating Adversarial Robustness of VR Cybersickness Detection and Mitigation

2025-12-18

AI Summary

Existing deep learning–based motion sickness detection and adaptive mitigation systems are vulnerable to adversarial attacks, leading to false detections, erroneous interventions, and degraded immersion; moreover, no open-source, real-time, end-to-end robustness evaluation platform exists. Method: We propose the first open-source VR testing platform integrating real-world eye-tracking and motion sensor inputs, a dynamic visual tunneling mitigation mechanism, and multiple adversarial attack methods (MI-FGSM, PGD, C&W), implemented in Unity with HTC Vive Pro Eye hardware. Contribution/Results: Experiments demonstrate that C&W attacks degrade Transformer-based detection accuracy by 5.94×, and all attacks successfully disable the mitigation functionality. The platform establishes a reproducible robustness benchmark, addressing a critical gap in standardized evaluation tools for VR motion sickness systems.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Deep learning (DL)-based automated cybersickness detection methods, along with adaptive mitigation techniques, can enhance user comfort and interaction. However, recent studies show that these DL-based systems are susceptible to adversarial attacks; small perturbations to sensor inputs can degrade model performance, trigger incorrect mitigation, and disrupt the user's immersive experience (UIX). Additionally, there is a lack of dedicated open-source testbeds that evaluate the robustness of these systems under adversarial conditions, limiting the ability to assess their real-world effectiveness. To address this gap, this paper introduces Adversarial-VR, a novel real-time VR testbed for evaluating DL-based cybersickness detection and mitigation strategies under adversarial conditions. Developed in Unity, the testbed integrates two state-of-the-art (SOTA) DL models: DeepTCN and Transformer, which are trained on the open-source MazeSick dataset, for real-time cybersickness severity detection and applies a dynamic visual tunneling mechanism that adjusts the field-of-view based on model outputs. To assess robustness, we incorporate three SOTA adversarial attacks: MI-FGSM, PGD, and C&W, which successfully prevent cybersickness mitigation by fooling DL-based cybersickness models' outcomes. We implement these attacks using a testbed with a custom-built VR Maze simulation and an HTC Vive Pro Eye headset, and we open-source our implementation for widespread adoption by VR developers and researchers. Results show that these adversarial attacks are capable of successfully fooling the system. For instance, the C&W attack results in a $5.94x decrease in accuracy for the Transformer-based cybersickness model compared to the accuracy without the attack.

Problem

• Evaluates adversarial robustness of VR cybersickness detection and mitigation systems.
• Addresses lack of open-source testbeds for assessing real-world effectiveness under attacks.
• Investigates how adversarial perturbations degrade model performance and disrupt user experience.

Innovation

• Open-source VR testbed for adversarial robustness evaluation
• Integrates DeepTCN and Transformer models with visual tunneling
• Implements MI-FGSM, PGD, and C&W attacks to assess vulnerabilities