A crowdsourcing-based QoE evaluation of an immersive VR autonomous driving experience

Abstract

Due to COVID-19, crowdsourcing has gained momentum as an alternative methodology for continuing research and for Quality of Experience (QoE) assessment. Employing this approach, we remotely evaluated the user perceived QoE of two different visual rendering formats as part of an Autonomous Vehicles (AVs) simulation. The aim was to investigate the participant’s QoE when testing AV technology in distinct visual rendering qualities (lowpoly vs high-poly) of an online streamed 360° car riding experience. In addition, a scoring model based on the expected reliability of each level of the remote assessment was designed. Findings suggest that the consumer’s preferences towards the adoption of AV technology is highly determined by the system and human effects on Influence Factors (IFs). Moreover, the adequacy of reliability into a mathematical model is highlighted as a potential turning point for QoE assessment, by carrying out the evaluation tasks from the laboratory environment into the internet, particularly relevant in pandemic times.