Easy-to-use, general, and accurate multi-Kinect calibration and its application to gait monitoring for fall prediction

Falls are the most-common causes of unintentional injury and death in older adults. Many clinics, hospitals, and health-care providers are urgently seeking accurate, low-cost, and easy-to-use technology to predict falls before they happen, e.g., by monitoring the human walking pattern (or “gait”). Despite the wide popularity of Microsoft's Kinect and the plethora of solutions for gait monitoring, no strategy has been proposed to date to allow non-expert users to calibrate the cameras, which is essential to accurately fuse the body motion observed by each camera in a single frame of reference. In this paper, we present a novel multi-Kinect calibration algorithm that has advanced features when compared to existing methods: 1) is easy to use, 2) it can be used in any generic Kinect arrangement, and 3) it provides accurate calibration. Extensive real-world experiments have been conducted to validate our algorithm and to compare its performance against other multi-Kinect calibration approaches, especially to show the improved estimate of gait parameters. Finally, a MATLAB Toolbox has been made publicly available for the entire research community.

[1]  K. S. Arun,et al.  Least-Squares Fitting of Two 3-D Point Sets , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Michael F Ballesteros,et al.  Recent trends in mortality from unintentional injury in the United States. , 2006, Journal of safety research.

[3]  Petros Daras,et al.  Estimating human motion from multiple Kinect sensors , 2013, MIRAGE '13.

[4]  François Charpillet,et al.  A gait analysis method based on a depth camera for fall prevention , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[5]  Georgy L. Gimel'farb,et al.  Multi-Kinect scene reconstruction: Calibration and depth inconsistencies , 2013, 2013 28th International Conference on Image and Vision Computing New Zealand (IVCNZ 2013).

[6]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[7]  Lois A Fingerhut,et al.  Deaths: injuries, 2002. , 2006, National vital statistics reports : from the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System.

[8]  Leigh Blizzard,et al.  A population-based study of sensorimotor factors affecting gait in older people. , 2008, Age and ageing.

[9]  Jean-Yves Bouguet,et al.  Camera calibration toolbox for matlab , 2001 .

[10]  Terry Caelli,et al.  Multi-kinect skeleton fusion for physical rehabilitation monitoring , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[11]  Brian Caulfield,et al.  SHIMMER: A new tool for temporal gait analysis , 2009, 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[12]  P O Riley,et al.  Effect of age on lower extremity joint moment contributions to gait speed. , 2001, Gait & posture.

[13]  Sander Oude Elberink,et al.  Accuracy and Resolution of Kinect Depth Data for Indoor Mapping Applications , 2012, Sensors.

[14]  Marcus A. Magnor,et al.  Markerless Motion Capture using multiple Color-Depth Sensors , 2011, VMV.

[15]  Marjorie Skubic,et al.  Unobtrusive, Continuous, In-Home Gait Measurement Using the Microsoft Kinect , 2013, IEEE Transactions on Biomedical Engineering.

[16]  Ran Gilad-Bachrach,et al.  Full body gait analysis with Kinect , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[17]  Ralf Reulke,et al.  CALIBRATION OF A MULTIPLE STEREO AND RGB-D CAMERA SYSTEM FOR 3D HUMAN TRACKING , 2014 .

[18]  Fabio Morbidi,et al.  Easy-to-Use and Accurate Calibration of RGB-D Cameras from Spheres , 2013, PSIVT.

[19]  Joris De Schutter,et al.  Multi RGB-D camera setup for generating large 3D point clouds , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[20]  Widyawan,et al.  Indoor human tracking application using multiple depth-cameras , 2012, 2012 International Conference on Advanced Computer Science and Information Systems (ICACSIS).

[21]  J. Stevens,et al.  The injury problem among older adults: mortality, morbidity and costs. , 2006, Journal of safety research.

[22]  Yeh-Liang Hsu,et al.  A Review of Accelerometry-Based Wearable Motion Detectors for Physical Activity Monitoring , 2010, Sensors.