Deep Learning for Content-based Personalized Viewport Prediction of 360-Degree VR Videos
This addresses the problem of personalized viewport prediction for VR video users, representing an incremental improvement over existing methods.
The paper tackles head movement prediction for 360-degree VR videos by introducing a deep learning network that uses both position data and video frame content, achieving a 16.1% improvement in prediction accuracy over a baseline using only position data.
In this paper, the problem of head movement prediction for virtual reality videos is studied. In the considered model, a deep learning network is introduced to leverage position data as well as video frame content to predict future head movement. For optimizing data input into this neural network, data sample rate, reduced data, and long-period prediction length are also explored for this model. Simulation results show that the proposed approach yields 16.1\% improvement in terms of prediction accuracy compared to a baseline approach that relies only on the position data.