Unlock instant, AI-driven research and patent intelligence for your innovation.

Kinect sensor based two-hand tracking method and device

A tracking device and sensor technology, applied in the field of gesture recognition and hand tracking, can solve the problems of high computational complexity, unsatisfactory recognition and tracking effects, too many restrictions on user posture, etc., and achieve the effect of low computational complexity

Active Publication Date: 2014-01-22
SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
View PDF3 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the prior art, there is a method of using the depth image and color image provided by Kinect to build a 3D model of both hands and track them, which can accurately locate detailed information such as the joints of both hands, but this method has a very high computational complexity, even when using GPU ( Graphic Processing Unit) acceleration can not achieve real-time tracking
Because Kinect itself has the function of extracting and tracking the human skeleton, it is also useful in the prior art to find the hand node on the skeleton to extract the hand position so as to track the hands, but this method requires the user's posture to be a standard sitting posture or Standing posture, too many restrictions on user posture, and the recognition and tracking effects are not ideal

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Kinect sensor based two-hand tracking method and device
  • Kinect sensor based two-hand tracking method and device
  • Kinect sensor based two-hand tracking method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] A method for tracking both hands based on Kinect, comprising:

[0047] S1 video acquisition step, obtains the color video stream and the depth video stream with the same resolution and frame rate from the Kinect sensor;

[0048] S2 Detection step 1, detecting the first hand that makes the initial gesture from the images of the acquired color video stream and depth video stream;

[0049] S3 single-hand tracking step, using the position information and size information of the first hand in the previous frame or more than two frames to locate and track the first hand;

[0050] S4 detection step 2, using the position information and size information of the first hand to detect the second hand;

[0051] S5 is a two-hands tracking step, tracking the detected hands.

[0052] The step S2 further includes: S2-1 sample training step; S2-2 mode selection step; S2-3 initial gesture determination step.

[0053] In the step S2-1, the SVM classifier is selected to learn and train the...

Embodiment 2

[0062] Such as figure 1 Shown is the flow chart of the hands tracking of this embodiment, including:

[0063] Step 1) Get the video stream. For example, the Kinect sensor is used to obtain a color video stream and a depth video stream with a resolution of 640*480 and a frame rate of 30fps.

[0064] Step 2) Detect the first hand. That is, detect the first hand from the acquired color map and depth map. Specifically, according to the depth information and skin color information or shape information, the nearest large enough target object is used as the first hand.

[0065] Step 3) Determine whether the initial gesture is detected. That is, it is judged whether the first hand found is making the initial gesture. When the hand stretches forward and the distance from the body is above the threshold d, it is determined that the hand makes a valid initial gesture, and the tracking starts. Mark this hand as Hand1, and store its three-dimensional position information into the tra...

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

No PUM Login to View More

Abstract

The invention provides a Kinect sensor based two-hand tracking method and device. The tracking method comprises steps as follows: S1, video acquisition; S2, detection step I; S3, tracking of a single hand; S4, detection step II; and S5, tracking of two hands: the detected two hands are tracked. When the second hand is detected, information of the first hand is used, so that the tracking method can rapidly and accurately track movement of the two hands of a user, and the calculation complexity is low.

Description

technical field [0001] The invention relates to gesture recognition and hands-tracking in the field of human-computer interaction, in particular to a method and device for hands-tracking based on a Kinect sensor. Background technique [0002] With the rapid development of computer technology and the advancement of people's scientific and technological concepts, users have put forward higher and higher requirements for the perception ability of computers. The traditional human-computer interaction mode mainly relies on the physical keyboard and mouse. Its relatively simple text-based input method can no longer meet people's needs, and the new human-computer interaction mode that pays more attention to "people-oriented" breaks the shackles of the traditional mode. , transforming the input method into a richer and more natural form such as images and sounds, which greatly improves the user experience. In recent years, people have made great progress in research fields such as ...

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
no application Login to View More
IPC IPC(8): G06T7/20G06K9/66
Inventor 朱艳敏袁博
Owner SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV