3D Posture Estimation from 2D Posture Data for Construction Workers

Construction workers’ behaviour is important for safety, health and productivity management. Workers’ 3D postures are the data foundation of their behaviours. This paper established a preliminary 3D posture dataset of construction tasks and provided a 3D posture estimation method based on 2D joint locations. The results showed that the method could estimate 3D postures accurately and timely. The mean joint error and estimation time of each frame were 1.10 cm and 0.12 ms respectively. This method makes it possible to estimate construction workers’ 3D postures from construction site images and contributes to a databased construction workers’ behaviour management.

[1]  Hong Zhang,et al.  Wearable IMU-based real-time motion warning system for construction workers' musculoskeletal disorders prevention , 2017 .

[2]  SangUk Han,et al.  Tracking-based 3D human skeleton extraction from stereo video camera toward an on-site safety and ergonomic analysis , 2016 .

[3]  W. P. Neumanna,et al.  A posture and load sampling approach to determining low-back pain risk in occupational settings , 2000 .

[4]  Yantao Yu,et al.  An experimental study of real-time identification of construction workers' unsafe behaviors , 2017 .

[5]  Heng Li,et al.  Ergonomic posture recognition using 3D view-invariant features from single ordinary camera , 2018, Automation in Construction.

[6]  Frédéric Bosché,et al.  Musculoskeletal disorders in construction: A review and a novel system for activity tracking with body area network. , 2016, Applied ergonomics.

[7]  Changbum R. Ahn,et al.  Fall risk analysis of construction workers using inertial measurement units: validating the usefulness of the postural stability metrics in construction , 2016 .

[8]  Xiaochun Luo,et al.  Joint-Level Vision-Based Ergonomic Assessment Tool for Construction Workers , 2019, Journal of Construction Engineering and Management.

[9]  Chang Liu,et al.  Evaluation of Classical Operators and Fuzzy Logic Algorithms for Edge Detection of Panels at Exterior Cladding of Buildings , 2019, Buildings.

[10]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[11]  Yaser Sheikh,et al.  OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[13]  JoonOh Seo,et al.  Construction Operation Simulation Reflecting Workers’ Muscle Fatigue , 2015 .

[14]  Yantao Yu,et al.  Estimating Construction Workers' Physical Workload by Fusing Computer Vision and Smart Insole Technologies , 2018, Proceedings of the 35th International Symposium on Automation and Robotics in Construction (ISARC).

[15]  SangUk Han,et al.  An Automated Biomechanical Simulation Approach to Ergonomic Job Analysis for Workplace Design , 2015 .

[16]  G. Legnani,et al.  A homogeneous matrix approach to 3D kinematics and dynamics — I. Theory , 1996 .

[17]  Peter E.D. Love,et al.  A deep hybrid learning model to detect unsafe behavior: Integrating convolution neural networks and long short-term memory , 2018 .

[18]  Jun Qiu,et al.  Construction worker's awkward posture recognition through supervised motion tensor decomposition , 2017 .

[19]  Xiaochun Luo,et al.  Capturing and Understanding Workers’ Activities in Far‐Field Surveillance Videos with Deep Action Recognition and Bayesian Nonparametric Learning , 2018, Comput. Aided Civ. Infrastructure Eng..

[20]  Xiaochun Luo,et al.  Towards efficient and objective work sampling: Recognizing workers' activities in site surveillance videos with two-stream convolutional networks , 2018, Automation in Construction.

[21]  Yaser Sheikh,et al.  OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  Fei Dai,et al.  Risk Assessment of Work-Related Musculoskeletal Disorders in Construction: State-of-the-Art Review , 2015 .

[23]  James J. Little,et al.  A Simple Yet Effective Baseline for 3d Human Pose Estimation , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[24]  Perry Forsythe,et al.  Evaluation of Terrestrial and Mobile Scanner Technologies for Part-Built Information Modeling , 2018, Journal of Construction Engineering and Management.

[25]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[26]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[27]  Yantao Yu,et al.  Visualization technology-based construction safety management: A review , 2017 .

[28]  Hong Zhang,et al.  Development of ergonomic posture recognition technique based on 2D ordinary camera for construction hazard prevention through view-invariant features in 2D skeleton motion , 2017, Adv. Eng. Informatics.

[29]  H. W. Heinrich,et al.  Industrial accident prevention : a safety management approach , 1980 .

[30]  Mohammad Nahangi,et al.  Level-of-Expertise Classification for Identifying Safe and Productive Masons , 2017 .