In recent years, as the demographic profile of society continues to shift towards an aging population, there has been a concomitant shortage of caregivers, leading to an increase in the demand for elderly care. The accurate assessment of the health status of the elderly and the provision of appropriate care necessitate the timely recognition and analysis of human activities. To address this challenge, we propose a continuous human activity recognition system that generates a 3D human skeleton model, utilizes joint angles to perform daily life activity recognition, and infer similarities in movements across various body parts. The proposed system generates a 3D human skeleton model using depth information obtained from multiple range-based depth cameras and extracts human joint angles on the basis of this model. Moreover, it utilizes time-series joint angle data to continuously recognize actions and estimate the similarity of movements across various body parts. To validate the efficacy of the proposed system, comprehensive verification experiments were conducted using real-world data.