Human movement expression method

Abstract

The invention discloses a representation method of human motion, which maps the human motion to a low-dimensional embedding space through nonlinear dimension reduction; models the data after dimension reduction with a linear time series model. Among them, the human motion is mapped to a low-dimensional embedded attitude space through nonlinear dimensionality reduction; the realization is: z=(x1+jy1, x2+jy2,...,xM+jyM)T. Model the dimensionally reduced data with a linear time series model; implementation steps: motion modeling based on a linear time series model; establish a p-order autoregressive model (AR) including; a p-order autoregressive model (AR) AR (p) has the following parameters: coefficient matrix Ak∈Rm×m, the parameter v introduced to ensure that the mean value of the dynamic process is non-zero, and the covariance matrix Q of Gaussian white noise. Let Ak be a diagonal matrix, then z(t) Each component of is independent; Given two autoregressive models (AR) A=[v, A1, A2,..., Ap] and A'=[v', A1', A2',..., Ap ’], then the distance metric D(A, A’)=‖A-A′∥F, where ‖·‖F represents the F-norm of the matrix.

Description

Technical field [0001] The invention relates to a representation method of human motion, belonging to the field of computer vision and the field of intelligent analysis of video content. Background technique [0002] In the field of computer vision and video content intelligent analysis, human motion analysis has become a very important and cutting-edge research topic [1-7] . In human motion analysis, human motion detection and tracking belong to low-level processing in vision, while human motion expression and understanding belong to high-level processing. Human movement expression and understanding play a vital role in video intelligent surveillance and other application fields. [0003] In the past ten years, many ways of expressing people's movements have emerged [1-6] . Among them, most research work extracts static information in each frame of image or motion information of adjacent frames to represent human motion. Efros et al [1] The optical flow is used to express people...

AI Technical Summary

Problems solved by technology

[0008] The main purpose of the present invention is to provide a method for expressing human motion to solve the current problem in the field of computer vision and intelligent analysis of video content that the data dimension before analysis using a linear time series model is particularly high. More training data is required, and the estimation of parameters is not accurate enough