Short-Term Trajectory Prediction for Full-Immersive Multiuser Virtual Reality with Redirected Walking

TL;DR

This paper compares RNN variants for predicting VR user trajectories under redirected walking, demonstrating that GRUs outperform LSTMs, and that incorporating virtual context improves accuracy in multi-user scenarios.

Contribution

It introduces the use of GRU networks for VR trajectory prediction, showing they outperform LSTMs, and demonstrates the benefit of including virtual environment context and scalability to multi-user systems.

Findings

GRUs outperform LSTMs in trajectory prediction accuracy.

Including virtual environment context improves prediction.

Prediction models scale well to multi-user VR scenarios.

Abstract

Full-immersive multiuser Virtual Reality (VR) envisions supporting unconstrained mobility of the users in the virtual worlds, while at the same time constraining their physical movements inside VR setups through redirected walking. For enabling delivery of high data rate video content in real-time, the supporting wireless networks will leverage highly directional communication links that will "track" the users for maintaining the Line-of-Sight (LoS) connectivity. Recurrent Neural Networks (RNNs) and in particular Long Short-Term Memory (LSTM) networks have historically presented themselves as a suitable candidate for near-term movement trajectory prediction for natural human mobility, and have also recently been shown as applicable in predicting VR users' mobility under the constraints of redirected walking. In this work, we extend these initial findings by showing that Gated Recurrent Unit (GRU) networks, another candidate from the RNN family, generally outperform the traditionally utilized LSTMs. Second, we show that context from a virtual world can enhance the accuracy of the prediction if used as an additional input feature in comparison to the more traditional utilization of solely the historical physical movements of the VR users. Finally, we show that the prediction system trained on a static number of coexisting VR users be scaled to a multi-user system without significant accuracy degradation.