东南大学学报（英文版）2025，Vol.41Issue(4)：446-456,11.DOI:10.3969/j.issn.1003-7985.2025.04.006
基于计算机视觉的施工行为实时追踪与虚拟仿真研究
Computer vision-based real-time tracking and virtual simulation of construction behavior
摘要
Abstract
To improve the safety of construction workers and help workers remotely control humanoid robots in construc-tion,this study designs and implements a computer vision-based virtual construction simulation system.For this pur-pose,human skeleton motion data are collected using a Ki-nect depth camera,and the obtained data are optimized via abnormal data elimination,smoothing,and normalization.MediaPipe extracts three-dimensional hand motion coordi-nates for accurate human posture tracking.Blender is used to build a virtual worker and site model,and the virtual worker motion is controlled based on the quaternion inverse kinematics algorithm while limiting the joint angle to en-hance the authenticity of motion simulation.Experimental results show that the system frame rate is stable at 60 frame/s,end-to-end delay is less than 20 ms,and virtual task comple-tion time is close to the real scene,verifying its engineering applicability.The proposed system can drive virtual work-ers to perform tasks and provide technical support for con-struction safety training.关键词
虚拟施工/计算机视觉/运动控制/人体运动姿态追踪/模拟仿真Key words
virtual construction/computer vision/motion control/human motion posture tracking/simulation分类
信息技术与安全科学引用本文复制引用
朱力,田芮竹,郭甲超,李佳欢,刘伟,赵冠远..基于计算机视觉的施工行为实时追踪与虚拟仿真研究[J].东南大学学报（英文版）,2025,41(4):446-456,11.基金项目
The Eighth National"Ten Thousand Talents Plan for Top Young Talents"of China,the National Natural Science Foun-dation of China(No.52478117,52378120). （No.52478117,52378120）
