Improved sign language recognition research using depth image information and SLVW
Yang Qua · Computer Engineering and Applications Journal · 2013
In order to realize the accurate recognition of manual alphabets in the sign language video, this paper presents an improved algorithm based on DI_CamShift(Depth Image CamShift)and SLVW(Sign Language Visual Word). It uses Kinect as the sign language video capture device to obtain both of the color video and depth image information of sign language gestures.The paper calculates spindle direction angle and mass center position of the depth images to adjust the search window and for gesture tracking. An Ostu algorithm based on depth integral image is used to gesture segmentation, and the SIFT features are extracted. It builds the SLVW bag of words as the feature of sign language and uses SVM for recognition. The best recognition rate of single manual alphabet can reach 99.87%, and the average recognition rate is 96.21%.