Dynamic Gesture Recognition Based on Depth Information

De GU; Ji LI

Human machine interaction by body language is becoming popular recently. With the help of 3D camera, video stream with depth information provides more detailed data to describe a movement. This paper proposed an algorithm to recognize dynamic gestures. Data preparation is needed first to eliminate some distractions. Then the start and the end of a possible meaningful gesture should be made clear. Finally, Dynamic Time Warping (DTW) is employed to calculate the similarity between a sample stream and the template. The test results show that the algorithm works well.

Select Volume / Issues:

• Volume 6,Issue 4, July 2015

Year:

2015

Type of Publication:

Article

Keywords:

Depth Information; Dynamic Time Warping; Gesture; Recognition

Journal:

IJECCE

Volume:

6

Number:

4

Pages:

493-496

Month:

July