Leveraging Digital Perceptual Technologies for Remote Perception and Analysis of Human Biomechanical Processes: A Contactless Approach for Workload and Joint Force Assessment

TL;DR

The paper addresses the need for non-invasive biomechanical analysis in real-world industrial settings by introducing a markerless computer vision framework that integrates with OpenSim. It combines CNN-based 2D pose estimation, DLT/SVD-based 3D triangulation, SMPL-X volumetric body modeling, LSTM-driven data augmentation to a Biomech-57 marker set, and OpenSim inverse kinematics to estimate joint angles and anthropometrics from multi-camera video. Key contributions include achieving joint-angle accuracy within a few degrees of marker-based systems and weight/height estimations with modest errors, along with comprehensive validation against gold-standard measures in a real-world task suite. The approach enables scalable, non-invasive workload and exoskeleton analysis in industry, with practical implications for ergonomic assessment, rehabilitation, and return-to-work strategies, while acknowledging limitations in occlusion, dataset size, and real-time performance with future improvements proposed.

Abstract

This study presents an innovative computer vision framework designed to analyze human movements in industrial settings, aiming to enhance biomechanical analysis by integrating seamlessly with existing software. Through a combination of advanced imaging and modeling techniques, the framework allows for comprehensive scrutiny of human motion, providing valuable insights into kinematic patterns and kinetic data. Utilizing Convolutional Neural Networks (CNNs), Direct Linear Transform (DLT), and Long Short-Term Memory (LSTM) networks, the methodology accurately detects key body points, reconstructs 3D landmarks, and generates detailed 3D body meshes. Extensive evaluations across various movements validate the framework's effectiveness, demonstrating comparable results to traditional marker-based models with minor differences in joint angle estimations and precise estimations of weight and height. Statistical analyses consistently support the framework's reliability, with joint angle estimations showing less than a 5-degree difference for hip flexion, elbow flexion, and knee angle methods. Additionally, weight estimation exhibits an average error of less than 6 % for weight and less than 2 % for height when compared to ground-truth values from 10 subjects. The integration of the Biomech-57 landmark skeleton template further enhances the robustness and reinforces the framework's credibility. This framework shows significant promise for meticulous biomechanical analysis in industrial contexts, eliminating the need for cumbersome markers and extending its utility to diverse research domains, including the study of specific exoskeleton devices' impact on facilitating the prompt return of injured workers to their tasks.

Figures (15)

• Figure 1: An example of CNN architectureDemertzis2023
• Figure 2: A typical RNN architectureBanerjee2019
• Figure 3: The structure of the developed framework
• Figure 4: The landmarks (keypoints) representation Body_25B model used in current studycao2017realtime.
• Figure 5: A schematic of part affinity fields (PAFs) with the input image to the parsing result.