Method, system and device for classifying sports behavior patterns

Abstract

The invention discloses a motion behavior pattern classification method, system and device, and relates to the field of human motion recognition. The method includes: S1, determining a candidate motion behavior pattern, the candidate motion behavior pattern including the motion behavior pattern to be classified; S2, acquiring a time series of behavior data about the candidate movement behavior pattern through a motion control sensor; S3 , establishing the LSTM motion behavior pattern classification model through the time series; S4, predicting the motion behavior pattern to be classified through the LSTM motion behavior pattern classification model, using an iterative process to compare the time series obtained using T in the S4 The prediction results and the prediction results obtained by using the T+Δt time series to obtain the final prediction results, where T 1 ≤T≤T 2 ,T 1 >0,T 2 > T 1 , the incremental step size is set to Δt, Δt>0. The technical solution of the invention can improve the accuracy while reducing the classification delay of the motion behavior pattern.

Description

technical field [0001] The invention relates to the field of human motion recognition, in particular to a method, system and device for classifying motion behavior patterns. Background technique [0002] The application of sensor-based motion recognition in modern society is becoming more and more common, especially with the emergence of a large number of consumer-grade VR devices such as Oculus and HTC Vive and the maturity of technology, mass consumers are increasingly interested in VR applications. The characteristic of immersive experience makes it have extraordinary potential in education, media, entertainment and other fields. It is an important function of VR applications to enable the experiencer to move freely in the virtual scene and ensure the naturalness of the movement to the greatest extent, and it is also a goal that researchers are currently working hard on. [0003] At present, the movement methods mainly used in VR virtual environment for interaction mainl...

AI Technical Summary

Problems solved by technology

However, how to implement a standard motion capture system (such as Vicon) that maps real-world motion to a virtual environment has always been a challenging topic, because it requires extra effort to install external equipment (professional cameras, tracking suits, etc.) ), which will bring huge production, time, and calculation overhead

Although some alternatives, such as Kinect, can achieve fast installation and accurate positioning, it will greatly limit the user's mobile space, and it is difficult to apply to outdoor scenes due to the influence of light, noise and other factors

Images