誤り耐性を有するマルチビュー映像ストリーミングに関する研究
誤り耐性を有するマルチビュー映像ストリーミングに関する研究
批准号:
12J01946
负责人:
劉 志
金额:
$1.86万
依托单位国家:
日本
项目类别:
Grant-in-Aid for JSPS Fellows
财政年份:
2012
资助国家:
日本
项目状态:
已结题
起止时间:
2012-04-01 至 2015-03-31
中文摘要
多视图被认为是下一代视觉通信技术,而多视图是自由视点电视等各种应用的基本问题之一。在自由视点视频传输中,用户可以选择自己喜欢的任意视角。为了提供自由的视点选择,服务器将根据用户的视图请求以纹理+深度格式发送最近的两个相邻视图,以便用户随后使用基于深度图像渲染(deep image based rendering, DIBR)技术合成所请求视角的视图。针对无线信道中容易出现突发损耗的问题,提出了采用多描述编码(multi - description coding, MDC)进行自由视点视频流编码后,通过两条路径传输视频内容。相关结果已被IEEE信号处理专题杂志接受。另外,电池寿命消耗对于移动用户来说至关重要,如何平衡能量消耗和接收到的视频质量成为一个严重的问题。我们研究了这个问题,并将结果发表在《网络与计算机应用杂志》上。HetNets是4G和5G的一个主要话题。如何在多小区(基站)组成的场景下传输视频将成为一个重要的问题。面对用户不同的视频请求,如何协调小区间的推理是一个挑战。一些初步结果已在IEEE VTC2014春季会议上发表,并正在提交延期。同时,我们也在寻找更多的多视角的应用,比如用于博物馆展览。
英文摘要
Multiview is expected to be the next generation visual communication and multi-view is one of the fundamental problems for various applications such as free viewpoint TV.In free viewpoint video transmission, users can select any view angle he prefers. To provide the free viewpoint selection, server will send the two nearest neighboring view in texture plus depth format upon user’s view request, so that the user can synthesize the view at the requested viewing angle afterwards using the depth image based rendering (DIBR) techniques. To handle the burst-loss prone wireless channels, we proposed to transmit the video contents over two paths after encoding them using the multiple descriptions coding (MDC) for the free viewpoint video streaming. The related results have been accepted to IEEE Journal of Selected Topics in Signal Processing.Orthogonally, battery life consumption is critical for mobile users, how to balance the energy consumption and the received video quality becomes a serious problem. We investigate this problem and publish the results in Journal of Network and Computer Applications.HetNets is a major topic in 4G and 5G. How to deliver the video in the scenarios composed of multiple cells (base station) will be become an important issue. The challenges exist in how to coordinate the intercell inferences given users’ diverse video requests. Some preliminary results have been published in IEEE VTC2014 spring and an extension is under submission.At the same time, we are looking for more applications of the multi-view such as for the museum exhibition.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
登录
查看更多内容
DOI:
10.1109/icc.2013.6655572
发表时间:
2013-06
期刊:
2013 IEEE International Conference on Communications (ICC)
影响因子:
--
作者:
[Zhi Liu;Yu Mao;N. Lu;Yusheng Ji;Xuemin Shen]
通讯作者:
Zhi Liu;Yu Mao;N. Lu;Yusheng Ji;Xuemin Shen
DOI:
10.1109/iccnc.2014.6785323
发表时间:
2014-04
期刊:
2014 International Conference on Computing, Networking and Communications (ICNC)
影响因子:
--
作者:
[Zhi Liu;Jingyun Feng;Yusheng Ji;Yongbing Zhang]
通讯作者:
Zhi Liu;Jingyun Feng;Yusheng Ji;Yongbing Zhang
DOI:
10.1109/vtcfall.2014.6966228
发表时间:
2014-12
期刊:
2014 IEEE 80th Vehicular Technology Conference (VTC2014-Fall)
影响因子:
--
作者:
[Ruijian An;Zhi Liu;Yusheng Ji]
通讯作者:
Ruijian An;Zhi Liu;Yusheng Ji
Multiple Description Coding of Free Viewpoint Video for Multi-Path Network Streaming
多路径网络流的自由视点视频的多重描述编码
DOI:
--
发表时间:
2012
期刊:
影响因子:
--
作者:
[Z. LIU, G. CHEUNG, J. CHAKARESKI, Y. JI]
通讯作者:
Y. JI
Multi-path Transmission for Picocell Edge Users in LTE-Advanced Heterogeneous Networks
LTE-Advanced 异构网络中 Picocell 边缘用户的多路径传输
DOI:
--
发表时间:
2015
期刊:
影响因子:
--
作者:
[Zhi Liu, Bo Gu, Lei Zhong, Yusheng Ji, Yohiaki Tanaka]
通讯作者:
Yohiaki Tanaka
共 13 条
Federated Reinforcement Learning Empowered Point Cloud Video Streaming
-
批准号:24K14927
-
项目类别:Grant-in-Aid for Scientific Research (C)
-
资助金额:$2.91万
-
财政年份:2024
-
负责人:劉 志
-
依托单位:
AI-empowered Point Cloud Video Streaming
-
批准号:20H04174
-
项目类别:Grant-in-Aid for Scientific Research (B)
-
资助金额:$10.98万
-
财政年份:2020
-
负责人:劉 志
-
依托单位:
海外基金