Deep learning–based video analytics demands high network bandwidth to ferry the large volume of data when deployed on the cloud. When incorporated at the edge side, only lightweight deep neural network (DNN) models are affordable due to computational constraint. In this article, a cloud–edge collaborative architecture is proposed combining edge-based inference with cloud-assisted continuous learning. Lightweight DNN models are maintained at the edge servers and continuously retrained with a more comprehensive model on the cloud to achieve high video analytics performance while reducing the amount of data transmitted between edge servers and the cloud. The proposed design faces the challenge of constraints of both computation resources at the edge servers and network bandwidth of the edge–cloud links. An accuracy gradient-based resource allocation algorithm is proposed to allocate the limited computation and network resources across different video streams to achieve the maximum overall performance. A prototype system is implemented and experiment results demonstrate the effectiveness of our system with up to 28.6% absolute mean average precision gain compared with alternative designs.

Ya Nan

Shiqi Jiang

Mo Li

ACM Transactions on Sensor Networks

ACM TRANSACTIONS ON SENSOR NETWORKS (TOSN) 是 ACM 在传感器网络跨学科领域的核心出版物，涵盖从信号处理、网络和协议、嵌入式系统、信息管理到分布式算法的广泛学科。它涵盖了引入新概念、技术、分析或架构的研究贡献，以及报告新工具和系统开发或具有高影响力、创新应用的经验和实验的应用贡献。这些交易特别关注对传感器网络系统方法的贡献以及基本贡献。

ACM Transactions on Sensor Networks (TOSN) is a key publication by ACM in the interdisciplinary field of sensor networks, covering a wide range of subjects from signal processing, networking and protocols, embedded systems, information management, to distributed algorithms. It accepts research contributions that introduce new concepts, technologies, analyses, or architectures, as well as reports on the development of new tools and systems, or empirical and experimental contributions with high impact and innovative applications. The journal particularly emphasizes contributions to sensor network system methodologies and foundational research.

《ACM Transactions on Sensor Networks》是ACM在传感器网络跨学科领域的重要出版物，涵盖信号处理、网络与协议、嵌入式系统、信息管理以及分布式算法等多个学科。期刊接受关于新概念、技术、分析或架构的研究贡献，以及新工具和系统开发的报告，或具有高影响力和创新应用的经验和实验贡献。特别关注对传感器网络系统方法的贡献及其基本研究。

Large-scale Video Analytics with Cloud–Edge Collaborative Continuous Learning

A cloud–edge collaborative architecture is proposed combining edge-based inference with cloud-assisted continuous learning to achieve high video analytics performance while reducing the amount of data transmitted between edge servers and the cloud.

Category	Quartile
计算机科学	4
计算机科学, 计算机信息系统	4
计算机科学, 电信学	4

Large-scale Video Analytics with Cloud–Edge Collaborative Continuous Learning

ivySCI AI Smartly Parses PDF, Answers Researchers' Questions, and Helps You Understand Papers in Seconds

Journal Info