A Chinese Sign Language Recognition Researh ...

An improved Depth Image CamShift (DI_CamShift) algorithm and SIFT-BoW feature are proposed to realize the accurate tracking of gestures in the sign language video and effective matching. First, it used Kinect to obtain the depth image information of sign language gestures. Second, it adjusted the search window by calculated spindle direction angle and mass center position of the depth images. Then the calculation of minimum depth information value in the search window was used to determine the target gesture area. Finally, SIFT-BoW is calculated for gesture matching and recognition. Experiments results show that the algorithm has good robustness and can effectively track and match the sign language gestures.