Vision-based two hand detection and tracking

In the field of human computer interaction (HCI), hand has been widely used as input device for natural interaction. However, during hand tracking, the continuously changing of hand shape and the interference from distractors (faces or hands) or occlusion reduce the robustness of man-machine alternation. In this paper, we use a web camera to detect two hands automatically and then track them stably in order to go against the problems mentioned above. At the first stage, we put forward a contour-based method to extract five fingertips which provide cues to locate initial hand position. At the second stage, CamShift is adopted to track the located hands. However, the means may easily lose the tracked objects due to its inadaptability to distractors and occlusion. Hence, we employ an improved Grey Model which is a good predictor of historical data to guide CamShift so as to achieve more accurate tracking. Experiments have been divided into two groups including the distractors tests and the occlusion tests. The convincing results illustrate the effectiveness of the proposed algorithm.

[1]  Lijun Xie,et al.  Visual Mouse: SIFT Detection and PCA Recognition , 2007, 2007 International Conference on Computational Intelligence and Security Workshops (CISW 2007).

[2]  Xiaojuan Wu,et al.  Research of a Real-time Hand Tracking Algorithm , 2005, 2005 International Conference on Neural Networks and Brain.

[3]  Ching-Chang Wong,et al.  Fuzzy tracking method with a switching grey prediction for mobile robot , 2001, 10th IEEE International Conference on Fuzzy Systems. (Cat. No.01CH37297).

[4]  Ahmet M. Kondoz,et al.  Two Hand Tracking Using Colour Statistical Model with the K-means Embedded Particle Filter for Hand Gesture Recognition , 2008, 2008 7th Computer Information Systems and Industrial Management Applications.

[5]  Miaolong Yuan,et al.  Robust hand tracking using a simple color classification technique , 2008, VRCAI '08.

[6]  Canhui Cai,et al.  Dual Searching Window Based Face Tracking , 2006, 2006 International Symposium on Intelligent Signal Processing and Communications.

[7]  Jianxin Zhang,et al.  Face Tracking with Occlusion , 2009, 2009 International Conference on Measuring Technology and Mechatronics Automation.

[8]  Guillaume-Alexandre Bilodeau,et al.  Face and Hands Detection and Tracking Applied to the Monitoring of Medication Intake , 2008, 2008 Canadian Conference on Computer and Robot Vision.

[9]  Paul A. Beardsley,et al.  Computer Vision for Interactive Computer Graphics , 1998, IEEE Computer Graphics and Applications.

[10]  Gary Bradski,et al.  Computer Vision Face Tracking For Use in a Perceptual User Interface , 1998 .

[11]  Jurij F. Tasic,et al.  Vision-based human-computer interface using hand gestures , 2007, Eighth International Workshop on Image Analysis for Multimedia Interactive Services (WIAMIS '07).

[12]  François Bérard,et al.  Bare-hand human-computer interaction , 2001, PUI '01.

[13]  Supavadee Aramvith,et al.  Improved face and hand tracking for sign language recognition , 2005, International Conference on Information Technology: Coding and Computing (ITCC'05) - Volume II.

[14]  Bo Yang,et al.  Target tracking using predicted Camshift , 2008, 2008 7th World Congress on Intelligent Control and Automation.

[15]  Yingyuan Xiao,et al.  Location Prediction for Tracking Moving Objects Based on Grey Theory , 2007, Fourth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD 2007).

[16]  Thomas Brown,et al.  Finger tracking for the Digital Desk , 2000, Proceedings First Australasian User Interface Conference. AUIC 2000 (Cat. No.PR00515).

[17]  Jinsong Xia,et al.  Face Tracking Based on Camshift Algorithm and Motion Prediction , 2009, 2009 International Workshop on Intelligent Systems and Applications.