SSE: A Smart Framework for Live Video Streaming based Alerting System

Abstract

Automated surveillance is always been a matter of curiosity due to its applications and the freedom through implicit monitoring. A smart implicit monitoring needs to be smarter with un-intervening inference-based classification, decision, and alerting processes. In this same sequence, detection and classification of unusual activities is the utmost curiosity among researchers. The entrance of Artificial Intelligence and the various computing ways (like Machine Learning and Deep Learning methods) of achieving it has been proven the most influential and promising computing revolution in the last decade. AI&ML-based object detection, segmentation, and identification has proven its vulnerability towards the achievement of these such goals and making computer vision smarter than ever before. In this paper, we are proposing a framework for an intelligent surveillance system based on AI&ML for video-based live surveillance. The proposed framework will provide a pathway to the intelligent system design for automated monitoring and alerting for unusual events based on detected objects. Basically, it would be a live streaming-based altering system.