Multi kinect cameras setup for skeleton based action recognition

In this research, we investigated and discussed several camera setup methods from different perspectives in order to exhibit advantages and shortcomings, as well as to reveal potentials of other camera setups in terms of image acquisition for action recognition system. Stability and accuracy of tracked skeletons of 10 actions from different views of camera settings are investigated. Experimental results indicates consistency of estimated skeleton lengths regardless of orientation placement of Kinect cameras. Accuracy of estimated skeleton data is less on actions with occluding body parts.

[1]  Charlie C. L. Wang,et al.  Improved Skeleton Tracking by Duplex Kinects: A Practical Approach for Real-Time Applications , 2013, J. Comput. Inf. Sci. Eng..

[2]  J. Muñoz,et al.  BKI: Brain Kinect Interface, a new hybrid BCI for rehabilitation , 2013, GFHEU.

[3]  K. Takahashi,et al.  Remarks on Real-Time 3D Human Body Posture Estimation using Multi-Camera System , 2007, IECON 2007 - 33rd Annual Conference of the IEEE Industrial Electronics Society.

[4]  Tao Wei,et al.  Kinect Skeleton Coordinate Calibration for Remote Physical Training , 2014, MMEDIA 2014.

[5]  Rob Aspin,et al.  Synchronization of Images from Multiple Cameras to Reconstruct a Moving Human , 2010, 2010 IEEE/ACM 14th International Symposium on Distributed Simulation and Real Time Applications.

[6]  Elena De Momi,et al.  Multi kinect people detection for intuitive and safe human robot cooperation in the operating room , 2013, 2013 16th International Conference on Advanced Robotics (ICAR).

[7]  Ivo Ihrke,et al.  A mobile system for multi-video recording , 2004 .

[8]  Piyush Rai,et al.  A Cost-effective Multiple Camera Vision System using FireWire Cameras and Software Synchronization , 2004 .

[9]  George A. Triantafyllidis,et al.  Synchronous Image Acquisition based on Network Synchronization , 2006, 2006 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'06).

[10]  M. Pollefeys,et al.  VIDEO SYNCHRONIZATION VIA SPACE-TIME INTEREST POINT DISTRIBUTION , 2004 .

[11]  Zeyu Chen,et al.  Gesture Recognition by Using Kinect Skeleton Tracking System , 2013, 2013 5th International Conference on Intelligent Human-Machine Systems and Cybernetics.

[12]  Rodrigo de Bem,et al.  Hardware and Software Infrastructure to Image Acquisition using Multiple Cameras , 2013 .

[13]  Luc Van Gool,et al.  ViRoom - Low Cost Synchronized Multicamera System and Its Self-calibration , 2002, DAGM-Symposium.

[14]  Antonios Gasteratos,et al.  Multi-camera 3D scene reconstruction from vanishing points , 2010, 2010 IEEE International Conference on Imaging Systems and Techniques.

[15]  Janusz Konrad,et al.  The Value of Multiple Viewpoints in Gesture-Based User Authentication , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[16]  Fei-Fei Li,et al.  Object discovery in 3D scenes via shape analysis , 2013, 2013 IEEE International Conference on Robotics and Automation.