CrowdVision: A Computing Platform for Video Crowdprocessing Using Deep Learning

CrowdVision: A Computing Platform for Video Crowdprocessing Using Deep Learning
复制标题

CrowdVision:使用深度学习进行视频众处理的计算平台

DOI:
10.1109/tmc.2018.2864212
复制
发表时间:
2019-07
期刊:
IEEE Transactions on Mobile Computing (TMC, CCF A类)
影响因子:
--
通讯作者:
Thomas La Porta
Thomas La Porta
中科院分区:
其他
文献类型:
--
作者:
Zongqing Lu;Kevin Chan;Shiliang Pu;Thomas La Porta

文献摘要

参考文献

被引文献

相似文献

智能手机等移动设备使用户能够以越来越高的速度生成和分享视频。在某些情况下,这些视频可能包含有价值的信息,可用于各种目的。然而,我们考虑的不是集中收集和处理视频以供信息检索,而是众包处理视频,即每个移动设备本地处理存储的视频。虽然移动设备的计算能力不断提高,但使用深度学习(即卷积神经网络)处理视频对移动设备来说仍然是一项艰巨的任务。为此,我们设计和构建了CrowdVision,这是一个计算平台,使移动设备能够利用云卸载,以分布式和节能的方式使用深度学习对视频进行众筹处理。CrowdVision可以在各种设置和不同网络连接下快速高效地处理带卸载的视频,并大大超过现有的计算卸载框架(例如,2倍的加速)。在这样做的过程中,CrowdVision解决了几个挑战:(I)如何利用深度学习的计算特性进行视频处理;(Ii)如何并行处理和卸载以实现加速;以及(Iii)如何通过确定正确的卸载时刻来优化运行时的时间和能量。
Mobile devices such as smartphones are enabling users to generate and share videos with increasing rates. In some cases, these videos may contain valuable information, which can be exploited for a variety of purposes. However, instead of centrally collecting and processing videos for information retrieval, we consider crowdprocessing videos, where each mobile device locally processes stored videos. While the computational capability of mobile devices continues to improve, processing videos using deep learning, i.e., convolutional neural networks, is still a demanding task for mobile devices. To this end, we design and build CrowdVision, a computing platform that enables mobile devices to crowdprocess videos using deep learning in a distributed and energy-efficient manner leveraging cloud offload. CrowdVision can quickly and efficiently process videos with offload under various settings and different network connections and greatly outperform the existing computation offload framework (e.g., with a 2× speed-up). In doing so, CrowdVision tackles several challenges: (i) how to exploit the characteristics of the computing of deep learning for video processing; (ii) how to parallelize processing and offloading for acceleration; and (iii) how to optimize both time and energy at runtime by just determining the right moments to offload.
DOI: 10.1145/2462456.2464440
发表时间: 2013-06
期刊: --
影响因子: --
作者:
P. Simoens;Yu Xiao;Padmanabhan Pillai;Zhuo Chen;Kiryong Ha;M. Satyanarayanan
通讯作者: P. Simoens;Yu Xiao;Padmanabhan Pillai;Zhuo Chen;Kiryong Ha;M. Satyanarayanan
DOI: 10.1145/3097895.3097903
发表时间: 2017-08
期刊: Proceedings of the Workshop on Virtual Reality and Augmented Reality Network
影响因子: --
作者:
Xukan Ran;Haoliang Chen;Zhenming Liu;Jiasi Chen
通讯作者: Xukan Ran;Haoliang Chen;Zhenming Liu;Jiasi Chen
DOI: 10.1109/infocom.2018.8486406
发表时间: 2018-04
期刊: IEEE INFOCOM 2018 - IEEE Conference on Computer Communications
影响因子: --
作者:
Zongqing Lu;Kevin S. Chan;T. L. Porta
通讯作者: Zongqing Lu;Kevin S. Chan;T. L. Porta
DOI: 10.1109/ipsn.2016.7460664
发表时间: 2016-04
期刊: 2016 15th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN)
影响因子: --
作者:
N. Lane;S. Bhattacharya;Petko Georgiev;Claudio Forlivesi;Lei Jiao;Lorena Qendro;F. Kawsar
通讯作者: N. Lane;S. Bhattacharya;Petko Georgiev;Claudio Forlivesi;Lei Jiao;Lorena Qendro;F. Kawsar
DOI: 10.1145/2632951.2632958
发表时间: 2014-08
期刊: --
影响因子: --
作者:
Cong Shi;Karim Habak;Pranesh Pandurangan;M. Ammar;M. Naik;E. Zegura
通讯作者: Cong Shi;Karim Habak;Pranesh Pandurangan;M. Ammar;M. Naik;E. Zegura