Deep-Framework: A Distributed, Scalable, and Edge-Oriented Framework for Real-Time Analysis of Video Streams

Abstract

Edge computing is the best approach for meeting the exponential demand and the real-time requirements of many video analytics applications. Since most of the recent advances regarding the extraction of information from images and video rely on computation heavy deep learning algorithms, there is a growing need for solutions that allow the deployment and use of new models on scalable and flexible edge architectures. In this work, we present Deep-Framework, a novel open source framework for developing edge-oriented real-time video analytics applications based on deep learning. Deep-Framework has a scalable multi-stream architecture based on Docker and abstracts away from the user the complexity of cluster configuration, orchestration of services, and GPU resources allocation. It provides Python interfaces for integrating deep learning models developed with the most popular frameworks and also provides high-level APIs based on standard HTTP and WebRTC interfaces for consuming the extracted video data on clients running on browsers or any other web-based platform.

MDED-Framework: A Distributed Microservice Deep-Learning Framework for Object Detection in Edge Computing

“…Sassu A. et al [ 17 ] proposed a deep-learning-based edge framework that can analyze multi-streams in real time. Docker-based services are structured to be processed independently, and two example applications are shown.…”

“…Docker-based services are structured to be processed independently, and two example applications are shown. While [ 17 ] focuses on improving the performance of CPUs and GPUs, the end goal of deep learning applications is to improve the accuracy of the model. In this paper, we present a model that is efficient in a distributed environment and also performs well on data not used for training, focusing on both processing speed and model accuracy.…”

Evaluation of capability maturity of digital transformation of manufacturing industry in Yangtze River Delta based on SDAF model

“…Smart Client is a new rich application supported by NETF ramework, integrating Windows and the Internet. The smart client application pattern combines the functionality and flexibility of the rich client pattern with the ease of deployment and stability of the browser-based pattern [2].By making full use of serviceoriented architecture.NET technology, XML Web Service technology, intelligent client mode, design mode, distributed mode and other modern software engineering methods are applied, a strong adaptability, evolvable, easy to maintain distributed application framework is established. The service-oriented smart client distributed application framework(Smart Client Service-Oriented Distributed Application Framework, SC -SODAF) model of working mechanism mainly provide offline application system infrastructure.…”