A Research on Real-Time Hand-Detection Technology Oriented to the Somatic Games
2013 ◽
Vol 748
◽
pp. 999-1002
◽
Keyword(s):
Hand-detection is a key technology to the somatic games. In this paper, we present a real-time hand-detection method based on Adaboost and skin-color characteristic. By processing the video frames with Adaboost classifier, we abstract the target regions which may contain the hand gestures. Then a filter based on skin color is proposed to select the correct regions. The best detection rate reaches above 89% with an acceptable failure rate and misjudgment rate. Experimental results show that this method is a lightweight and rapid approach to implement real-time hand detection in somatic games.
Keyword(s):
2013 ◽
Vol 302
◽
pp. 772-775
◽
Keyword(s):
2012 ◽
Vol 25
(7)
◽
pp. 1331-1337
◽
Keyword(s):
2010 ◽
Vol 130
(11)
◽
pp. 2039-2046
Keyword(s):