A real-time and low-cost hand tracking system

This paper presents a high performance and low-cost real-time hand tracking system which runs on general consumer electronics with very low CPU consumption. The hand tracker fuses motion feature with color feature to alleviate the drifting problem during tracking. Experimental results on two challenging video sequences show that the proposed hand tracking system outputs the state-of-the-art performance, and consumes only about 5% CPU resources of a low-spec smart TV while processing VGA resolution video at 25 frames per second.

[1]  Nong Sang,et al.  A low-cost hand gesture human-computer interaction system , 2012, 2012 IEEE International Conference on Consumer Electronics (ICCE).

[2]  Nong Sang,et al.  Detecting skin colors under varying illumination , 2011, International Symposium on Multispectral Image Processing and Pattern Recognition.

[3]  Nong Sang,et al.  Real-time skin color detection under rapidly changing illumination conditions , 2011, IEEE Transactions on Consumer Electronics.

[4]  Hongzhi Wang,et al.  Real-Time Tracking Combined with Object Segmentation , 2014, 2014 22nd International Conference on Pattern Recognition.

[6]  Yu Zhou,et al.  Similarity Fusion for Visual Tracking , 2015, International Journal of Computer Vision.

[7]  Changxin Gao,et al.  Robust Visual Tracking Using Exemplar-Based Detectors , 2017, IEEE Transactions on Circuits and Systems for Video Technology.

[8]  Nong Sang,et al.  Background subtraction using shape and colour information , 2010 .

[9]  Dorin Comaniciu,et al.  Kernel-Based Object Tracking , 2003, IEEE Trans. Pattern Anal. Mach. Intell..