Presentation
A PRISMA Review of Human Pose Estimation Models for Ergonomic Risk Assessment and Workplace Safety
SessionPoster Session 1
DescriptionThis study presents a systematic review and meta-analysis of human pose estimation (HPE) applications for ergonomic risk assessment and workplace safety, following the PRISMA framework. Thirty-eight studies were reviewed to assess the effectiveness, usability, and limitations of HPE models such as OpenPose, MediaPipe, and AlphaPose in monitoring posture and identifying risks in occupational environments. The review found that HPE models facilitate non-invasive, real-time tracking of skeletal movements, supporting ergonomic assessments in dynamic and unstructured settings. While promising, these models still face challenges related to occlusion, lighting variability, and joint-specific accuracy. A meta-analysis revealed consistent improvements in posture and reductions in musculoskeletal risk following ergonomic interventions. These findings suggest that integrating HPE with ergonomic evaluation tools can enhance workplace safety and worker well-being. Future research should prioritize improving model efficiency and incorporating real-time feedback to support proactive ergonomic interventions and widespread adoption in industrial settings.
Event Type
Poster
TimeTuesday, October 14th5:30pm - 6:30pm CDT
LocationRiverside East