Training Hidden Markov Model Structure with Genetic Algorithm for Human Motion Pattern Classification

Abstract

Physical exercise classification method by Hidden Markov Model(HMM) is considered in this study. The aim of this study is to discuss the availability of HMM based motion modeling in order to compare human skills. In this paper, a preprocessing technique for observed human motion by Self-Organizing Map(SOM) to label a motion characteristic is proposed. Then, HMM construction method by using Genetic Algorithm(GA) with Baum-Welch algorithm, modified crossover and mutation is introduced. Simulation studies are carried out for bat swing motions. It is shown that the proposed approach has an ability to recognize bat swing motions.

Efficient human motion capture data annotation via multi‐view spatiotemporal feature fusion

“…To solve this issue, Barbic et al [9] first utilised the probabilistic PCA to detect the boundary between consecutive actions and then divided the human motion into distinct high-level behaviours. Later, Manabe et al [25] employed the key-frame representations to segment the continuous observation data into discrete symbols. Nevertheless, these two approaches often require some prior knowledge to detect the motion variation or stipulate motion keys in advance, which are not immediately accessible to the novice users in practice.…”

“…Later, Manabe et al . [25] employed the key‐frame representations to segment the continuous observation data into discrete symbols. Nevertheless, these two approaches often require some prior knowledge to detect the motion variation or stipulate motion keys in advance, which are not immediately accessible to the novice users in practice.…”