Assessing Ergonomic Compliance in Industrial Environments with Markerless 3D Camera-Based Systems

In this research, we introduce a markerless, 3D camera-based system for ergonomic assessment in industrial settings, utilizing two standard RGB cameras and the OpenPose framework for real-time 2D skeleton mapping. This novel approach overcomes the limitations of traditional, subjective ergonomic assessment methods and sensor-based systems by offering objective posture evaluations without physical markers. The dual-camera setup enables accurate 3D reconstruction of worker movements. Our feasibility study validates the system using the Intra-class Correlation Coefficient (ICC), achieving a ground truth agreement among ergonomic experts with an ICC of 0.97, and the Pearson Correlation Coefficient (PCC), demonstrating a strong correlation (PCC of 0.81) between expert evaluations and our camera-based measurements for neck and upper limb flexion. This highlights the system's reliability and potential as a cost-effective solution in industrial ergonomics.