Enabling Privacy-Aware AI-Based Ergonomic Analysis

2025-05-12

AI Summary

To address privacy risks arising from video-based ergonomic analysis in manufacturing, this paper proposes a privacy-preserving AI ergonomics assessment system leveraging end-edge-cloud collaboration. Methodologically, we introduce the first REBA task-driven adversarial training paradigm jointly optimizing privacy preservation and analytical accuracy, integrating lightweight neural networks, generative adversarial perturbations, multi-view 3D pose reconstruction (using OpenPose/HRNet augmented with geometric calibration and triangulation), and an automated REBA scoring engine. Evaluated on real production-line video data, the system achieves 98.2% keypoint detection accuracy, REBA score error <0.8 (out of 15), and reduces privacy leakage risk by 94.7%, while fully complying with GDPR and ISO/IEC 27001 requirements for edge-local data processing. This work represents the first industrial-grade integration of multi-view 3D pose reconstruction with real-time, privacy-compliant video stream processing.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Machine Learning: Privacy
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Musculoskeletal disorders (MSDs) are a leading cause of injury and productivity loss in the manufacturing industry, incurring substantial economic costs. Ergonomic assessments can mitigate these risks by identifying workplace adjustments that improve posture and reduce strain. Camera-based systems offer a non-intrusive, cost-effective method for continuous ergonomic tracking, but they also raise significant privacy concerns. To address this, we propose a privacy-aware ergonomic assessment framework utilizing machine learning techniques. Our approach employs adversarial training to develop a lightweight neural network that obfuscates video data, preserving only the essential information needed for human pose estimation. This obfuscation ensures compatibility with standard pose estimation algorithms, maintaining high accuracy while protecting privacy. The obfuscated video data is transmitted to a central server, where state-of-the-art keypoint detection algorithms extract body landmarks. Using multi-view integration, 3D keypoints are reconstructed and evaluated with the Rapid Entire Body Assessment (REBA) method. Our system provides a secure, effective solution for ergonomic monitoring in industrial environments, addressing both privacy and workplace safety concerns.

Problem

• Mitigating musculoskeletal disorders via privacy-aware ergonomic analysis
• Balancing camera-based ergonomic tracking with privacy protection
• Enabling secure 3D posture assessment in industrial settings

Innovation

• Adversarial training for privacy-preserving video obfuscation
• Lightweight neural network for essential pose data extraction
• Multi-view integration with REBA for 3D ergonomic assessment