First and Second Order Dynamics in a Hierarchical SOM system for Action Recognition

TL;DR

This paper introduces a hierarchical SOM-based action recognition system that leverages joint position and dynamics data, demonstrating improved accuracy through the inclusion of movement dynamics and an attention mechanism.

Contribution

The novel system combines hierarchical Self-Organizing Maps with a supervised neural network, incorporating joint dynamics and an attention mechanism for enhanced action recognition.

Findings

Including dynamics improves recognition accuracy.

Attention mechanism focuses on key body parts.

System outperforms less sophisticated preprocessing methods.

Abstract

Human recognition of the actions of other humans is very efficient and is based on patterns of movements. Our theoretical starting point is that the dynamics of the joint movements is important to action categorization. On the basis of this theory, we present a novel action recognition system that employs a hierarchy of Self-Organizing Maps together with a custom supervised neural network that learns to categorize actions. The system preprocesses the input from a Kinect like 3D camera to exploit the information not only about joint positions, but also their first and second order dynamics. We evaluate our system in two experiments with publicly available data sets, and compare its performance to the performance with less sophisticated preprocessing of the input. The results show that including the dynamics of the actions improves the performance. We also apply an attention mechanism that focuses on the parts of the body that are the most involved in performing the actions.