Gesture prediction method and system

Abstract

The present invention provides a gesture prediction method and system, the method comprising: detecting the current gesture start coordinates of the gesture guide body, and sequentially recording the sequence of the current movement trajectory coordinates of the gesture guide body starting from the current gesture start coordinates ; Every first preset time interval, extract the current motion track feature vector of the gesture guide body according to the sequence of the current motion track coordinates of the recorded gesture guide body; use the pre-trained gesture recognition model to extract the current motion track of the gesture guide body Pattern matching is performed on the feature vectors, and the matched complete gesture identification is output; wherein, the pre-trained gesture recognition model is obtained according to the training of the pre-collected gesture feature vector sample set, and the pre-collected gesture feature vector sample set includes the complete motion track of the complete gesture according to the preset ratio The eigenvectors are several local motion trajectory eigenvectors composed of continuous eigenvalues ​​intercepted from the beginning. The gesture prediction method and system of the present invention can realize real-time recognition of gestures.

Description

technical field [0001] The invention relates to the technical field of pattern recognition, in particular to a gesture prediction method and system. Background technique [0002] With the wide application of computers, the traditional human-computer interaction method of manipulating physical buttons is not natural and convenient, and using human hands as a means of human-computer interaction is more natural, direct, and richer than other biological features, and can greatly improve the human-computer interaction. Enhance user experience. The initial gesture recognition mainly uses external devices to directly detect the angle and spatial position of the joints of the hand and arm, such as typical data gloves. Although these external devices improve the accuracy and stability of gesture recognition, they do not conform to gesture recognition. natural expression. So vision-based gesture recognition came into being. The vision-based gesture interaction method is a non-conta...

AI Technical Summary

Problems solved by technology

[0004] Based on this, it is necessary to provide a gesture prediction method and system for the technical problem that the current gesture recognition method can only recognize and give the recognition result after the gesture is over, which affects the real-time performance of gesture recognition

Images