Online Recognition of Actions Involving Objects

TL;DR

This paper introduces an online system for real-time recognition of object-involving actions, combining hierarchical neural networks and proximity measures to identify actions and objects simultaneously.

Contribution

It presents a novel multi-stream neural network approach that merges spatial trajectory analysis with object proximity detection for online action recognition.

Findings

Achieved excellent performance in real-time action recognition

Successfully integrated action and object identification

Demonstrated effectiveness in online processing environments

Abstract

We present an online system for real time recognition of actions involving objects working in online mode. The system merges two streams of information processing running in parallel. One is carried out by a hierarchical self-organizing map (SOM) system that recognizes the performed actions by analysing the spatial trajectories of the agent's movements. It consists of two layers of SOMs and a custom made supervised neural network. The activation sequences in the first layer SOM represent the sequences of significant postures of the agent during the performance of actions. These activation sequences are subsequently recoded and clustered in the second layer SOM, and then labeled by the activity in the third layer custom made supervised neural network. The second information processing stream is carried out by a second system that determines which object among several in the agent's vicinity the action is applied to. This is achieved by applying a proximity measure. The presented method combines the two information processing streams to determine what action the agent performed and on what object. The action recognition system has been tested with excellent performance.