Viewing safety is one of the main issues in viewing virtual reality (VR) content. In particular, VR sickness can occur when watching immersive VR content. To deal with viewing safety for VR content, objective assessment of VR sickness is of great importance. In this work, based on a deep generative model, we propose a novel objective VR sickness assessment (VRSA) network to automatically predict VR sickness. The proposed method takes into account motion patterns of VR videos in which exceptional motion is a critical factor inducing excessive VR sickness in human motion perception. The proposed VRSA network consists of two parts, the VR video generator and VR sickness score predictor. For the evaluation of VRSA performance, we performed comprehensive experiments with 360° videos (stimuli), corresponding physiological signals, and subjective questionnaires. We demonstrated that the proposed VRSA achieved a high correlation with human perceptual score for VR sickness....read more