Development of an occupancy measurement system for micro-zones within open office spaces based on multi-view multi-person 3D pose estimation

Abstract

A micro-zone constitutes the fundamental unit of open spaces. Occupancy information of micro-zones with functional attributes is critical for layout optimization, space management, and energy control in open-plan workplaces. However, existing occupancy measurement methods predominantly focus on macro-scales and commonly suffer from reliance on additional devices, rigid behavioral constraints, and limited access to authentic human occupancy data. To address these limitations and bridge the gap in micro-scale occupancy analysis, this study proposes a design framework for occupancy measurement systems to investigate functional zone utilization in open office spaces. The framework takes multi-view RGB images as input, applies 3D pose estimation to infer spatial positions of human joints, and further determines the occupancy state and metrics of the entire space and functional zones by analyzing overlap between projected joint coordinates and zone boundaries. To evaluate the feasibility of the framework, a prototype system is developed and deployed in a small-scale open office space, achieving a Precision of 100% and an F1 score of 89.19% in occupancy state measurement. These results demonstrate that the framework effectively supports occupancy surveys of functional zones and holds potential for real-world application. This study presents a low-cost, multi-scale occupancy measurement approach that innovatively introduces pose estimation technology into building occupancy investigation. While enabling holistic spatial occupancy perception, it fills the gap in functional zone analysis. This work provides a theoretical foundation for the development of CCTV-based occupancy measurement methods and offers data support for decision-making in the sustainable design and operation of indoor open office environments.