Pose Recognition and Intelligent Video Analytics in Smart Home Security: Integrating AI with IoT-Based Surveillance Systems

Modern surveillance is being transformed by the combination of artificial intelligence and the Internet of Things, especially with intelligent video analytics and pose recognition technologies. This paper will discuss how AI-powered pose recognition in the context of smart-home security systems can work by means of an econometric analysis of the data on how the Guard-N 4.0 platform operates. The study deals with one of the most critical problems that is the accomplishment of a balance between detection accuracy, latency, energy consumption, and bandwidth consumption in privacy-preserving settings. The aim is to measure the effect that the enablement of pose recognition and the quality of pose has on detection performance and system latency. The analytical methodology is a mix of a secondary data. analysis and econometric modelling, which is the Negative Binomial Fixed-Effects, Seemingly Unrelated Regressions, Difference-in-Differences, and Instrumental Variables. It was analyzed on a panel dataset of 18,450 device home day observations on the parameters: alert counts, latency and energy consumption.