Preliminary Validation of a Low-Cost Motion Analysis System Based on RGB Cameras to Support the Evaluation of Postural Risk Assessment

This paper introduces a low-cost and low computational marker-less motion capture system based on the acquisition of frame images through standard RGB cameras. It exploits open-source deep learning model CMU, from tf-pose-estimation project. Its numerical accuracy its usefulness for ergonomic assessment are evaluated by proper experiment, designed performed to: (1) compare data provided it with those collected golden system; (2) RULA scores obtained Vicon Nexus estimated video analysis, team three expert ergonomists. Tests have been conducted in standardized laboratory conditions involved total six subjects. Results suggest that proposed can predict angles good consistency give evidence about tool’s ergonomist.