An advanced QER selection algorithm for 360-degree VR video streaming

PubDate: November 2019

Teams: Yonsei University

Writers: A-young Kim, Eun-bin An, Kwang-deok Seo

PDF: An advanced QER selection algorithm for 360-degree VR video streaming

Abstract

The 360-degree virtual reality (VR) video streaming faces many challenging issues since VR video contains a large volume of data and requires streaming omnidirectional view. For consuming VR video in a reasonable quality, the streaming system is required to efficiently use bandwidth and reduce the computational complexity of the client device. Quality Emphasized Region (QER) based streaming has been introduced as a kind of viewport adaptive streaming scheme for reducing waste of bandwidth. This scheme requires QER selection for providing the adaptive quality of QER. We propose an advanced QER selection algorithm consisting of two stages. The first stage is the calculation of the Quality Emphasis Center distance which employs preprocessed QER_ID_MAP for reducing the computational complexity. As the second stage, we propose an adaptive adjustment of the signaling interval between signaling messages. The simulations show that the proposed algorithm results in reduction of computational complexity and bandwidth waste.