A motion gesture recognition method and device for catching a ball

Abstract

A motion posture recognition method and device for catching a ball, which obtains the three-dimensional coordinates of all identifiable reflective marking points in the motion space, and obtains a plurality of matching points on the catching ball from all reflective marking points, and then according to these matching The three-dimensional coordinates of the point are calculated to obtain the coordinates of the center of the catching ball, and then the motion posture information of the catching ball in the motion space is calculated. Since the matching method between the identifiable reflective marker points and the catching ball is added in the process of obtaining multiple matching points on the catching ball, an iterative algorithm for calculating the center coordinates of the catching ball is introduced in the process of obtaining the center coordinates of the catching ball. The misrecognition situation caused by the wrong matching of the reflective marking points and the increase of the coordinate error of the center of the ball is avoided, and it is beneficial to improve the recognition degree of the system for catching the ball. At the same time, when the method and device are adopted, fewer matching points on the catching ball can be used for motion posture recognition, thereby reducing the calculation amount, increasing the calculation speed, and bringing convenience to practical applications.

Description

technical field [0001] The invention relates to the technical field of optical motion capture, in particular to a motion gesture recognition method and device for catching a ball. Background technique [0002] In the existing optical motion capture system, generally by capturing the movement track of reflective markers (markers, which refer to markers covered with special reflective materials on the surface, common shapes are spherical and hemispherical, etc., are often used to capture moving objects) In order to realize the motion posture recognition and trajectory tracking of the target object. Usually, a plurality of reflective marks configured on the capture object will form a rigid body that can be captured (the rigid body here refers to the shape and size of the rigid body during motion and after the force is applied, and the relative position of the internal points An unchanging object is an ideal model), and each reflective mark corresponds to a certain target part ...

AI Technical Summary

Problems solved by technology

[0004] (1) When recognizing the catching ball in the form of a rigid body, since there are a lot of reflective stickers on the catching ball to configure a large number of reflective marks, in order to fully recognize the motion posture of the catching ball, in the later stage of posture recognition and trajectory tracking process , to traverse the three-dimensional coordinates of each reflective mark, the calculation is huge, which will greatly affect the real-time motion gesture recognition and trajectory tracking effect of the catching ball

[0005] (2) The structure of the catching ball has a difference in size. A ball with a relatively large diameter will cause unpredictable occlusion of the reflective sticker during motion, which will affect the capture process and bring poor motion gesture recognition effect.

Images