A Multidimensional Weighted 3D Dynamic Gesture Recognition Method
A technology of dynamic gestures and recognition methods, applied in the input/output of user/computer interaction, mechanical mode conversion, computer parts and other directions, and can solve the problem of low gesture recognition rate
Inactive Publication Date: 2017-01-11
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF8 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the existing gesture recognition system has the problem of low gesture recognition rate.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0090] Specific embodiments of the present invention will be described below in conjunction with the accompanying drawings, so that those skilled in the art can better understand the present invention. It should be noted that in the following description, when detailed descriptions of known functions and designs may dilute the main content of the present invention, these descriptions will be omitted here.
[0091] In this example, if figure 1 As shown, first perform gesture segmentation, such as figure 2 As shown, the joint point data of standard gestures are obtained from the continuous human action video (training video) provided by the image input device;
[0092] Track the positions of the six joint points of the left and right hands, left and right wrists, and left and right elbows, and detect whether the hand is in the extended state, that is, whether the Z coordinate of the hand is the smallest among the six tracked joint points. If it is not in the extended state, re...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a multi-dimensional weighted 3D dynamic gesture recognition method. In the training stage, firstly, the standard gesture is segmented to obtain the feature vector of the standard gesture, and then coordinate system conversion, normalization processing, smoothing processing, down-sampling and Differential processing is used to obtain a standard gesture feature vector set and the weights of each joint point and each dimension of the element, so as to construct a standard gesture sample library. In the recognition stage, the multi-dimensional weighted dynamic time warping algorithm is used to calculate the dynamic warping distance between the feature vector set of the gesture to be recognized and the feature vector set c=1,2,...,C of each standard gesture in the standard gesture sample library. When the (m,n)th element s(m,n) of the cost matrix C, consider the weights of the joint points and the weights of each dimension in the element, and remove the joint points and coordinate dimensions that do not contribute to gesture recognition, effectively It removes the interference caused by joint shaking and human error movements on gesture recognition, improves the anti-interference ability of the algorithm, and finally improves the accuracy and real-time performance of gesture recognition.
Description
technical field [0001] The invention belongs to the technical fields of pattern recognition and intelligent systems, computer vision and human-computer interaction, and more specifically relates to a multi-dimensional weighted 3D dynamic gesture recognition method. Background technique [0002] The field of human-computer interaction has experienced two revolutions. The first was the appearance of the mouse in 1983, which allowed people to enter the two-dimensional graphical interface based on the mouse from the one-dimensional command line based on the keyboard. The second is the emergence of touch screen technology, which integrates display and input, making people gradually familiar with the interactive mode of multi-touch. The development of touch screen technology has also led to the exploration and research of other new interactive methods and devices. [0003] It can be considered that the transformation of the human-computer interaction mode is closely related to th...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F3/01
Inventor 康波李云霞孙琴蔡会祥
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA