3D-LIVE: D4.2 Second report on the experimentationsand evaluations of the 3D-LIVE Tele-Immersive Environment
Simon Crowle, Marco Conte, Benjamin Poussard · ePrints Soton (University of Southampton) · 2015
This report presents the final experimental phase conducted as part of the 3D-LIVE project. Technical progress that lead to this final phase includes new developments and refinements to existing work based on lessons learned from previous experimental work (see deliverable D4.1). These early experimental outcomes guided subsequent enhancements to the 3D-LIVE scenarios and the updates to the system prototype. In this final phase of investigation (LIVE3), the experimental focus has been refined and concentrated on those aspects of the 3DLIVE UX that most closely aligned with the quality of tele-immersive experience in a shared, mixed reality environment (according to the recommendations received during the Interim Review Meeting). LIVE 3 experiments were carried out between M26 and M29 of the project. User groups were engaged in a series of experiments that were devised to capture data intended to provide evidence of (anticipated) improvements in the principal dimensions of an immersive UX. Overall, we successfully conducted a total of three trials (one per scenario), engaging 60 participants across 27 experimental runs in total. In our analysis of the experimental data captured, we compare and contrast the QoE and QoS data generated within and between user groups and scenarios, scoped with specific experimental objectives. As well as reporting on the main user trials, we additionally describe comparisons between high-end, immersive, 3D displays and an evaluation of a network sensitive, adaptive compression algorithm for full body reconstruction data (part of task T3.6). Our analysis reveals generally good to very good QoE responses from our participants in many aspects of their experience; our results provide validation of the 3DLIVE platform as a system capable of delivering a compelling, mixed reality TI experience. We identify specific QoS influences and game design aspects that impact positively or negatively to overall UX in this context and make recommendations for future improvements of the 3D-LIVE platform. In addition, we propose heuristics for mixed reality system design and implementation. Finally, we provide an analysis of the 3D-LIVE cocreation methodology and its positive contributions to the project, and offer suggestions for its application in other domains.