For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Segment Scheduling for Progressive Download-Based Multi-View Video Delivery under Successive View Switching
Takahito KITO Iori OTOMO Takuya FUJIHASHI Yusuke HIROTA Takashi WATANABE
IEICE TRANSACTIONS on Communications
Publication Date: 2018/04/01
Online ISSN: 1745-1345
Type of Manuscript: PAPER
Category: Multimedia Systems for Communications
multiview video, progressive download, segment scheduling, successive view switching,
Full Text: PDF(2.5MB)
>>Buy this Article
In conventional multiview video systems using progressive download, a user downloads videos of all viewpoints of one content to realize smooth view switching. This, however, increases the video traffic, and if the available download rate is low, the video quality suffers. Downloading only the desired viewpoint is one approach for reducing the traffic. However, in this case, playback stalls will occur after view switching. These stalls degrade the user's satisfaction for the application. In this paper, we aim at two objectives: 1) to achieve reduction in video traffic and 2) to minimize the number of playback stalls. To this end, we propose a new multiview video delivery scheme for progressive download. The main idea of the proposed scheme is that the user downloads a part of viewpoints only, which will be played back by the user with a high probability, to realize both traffic reduction and smooth view switching. In addition, we propose two download-scheduling algorithms to prevent playback stalls even at low download rates. The first algorithm prevents stalls in the cases with frequent view switching, such as zapping, while the second prevents stalls in gazing cases. Evaluations using a Joint Multiview Video Coding (JMVC) encoder and multiview video sequences show that our scheme achieves not only reduced video traffic but also decreased number of playback stalls, regardless of the user's view-switching model or download rate. In addition, we demonstrate that the proposed method does not cause playback stalls irrespective of high and low motion video contents.