Objective Evaluation of VR Sickness and Analysis of Its Relationship with VR Presence

Virtual Reality (VR) users often experience symptoms such as dizziness and nausea, known as VR sickness, which is a significant challenge to VR technology. VR Presence, indicating users’ immersion in VR environments, serves as a crucial metric for assessing VR product quality. To evaluate VR sickness, we propose an objective assessment model based on motion and visual information. Our model analyzes factors influencing VR sickness and extracts rotational, motion intensity, content and spatial features from VR visual contents. Based on the support vector regression algorithm, we construct a robust VR sickness assessment model. The model’s efficacy is validated on the KAIST and the VR-SP datasets, and the PLCC values are improved to 0.942 and 0.912, respectively. Finally, we explore the relationship between VR sickness and VR presence using VR-SP datasets. VR sickness and presence are positively correlated when the sickness score is below 0.646. When the sickness score was above 0.646, they were negatively correlated.