Hierarchical Self-Organizing Maps System for Action Classification

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding

Abstract

We present a novel action recognition system that is able to learn how to recognize and classify actions. Our system employs a three-layered hierarchy of Self-Organizing Maps together with a supervised neural network for labelling the actions. We have evaluated our system in an experiments consisting of ten different actions from a publicly available data set. The results are encouraging with 83% correctly classified actions based on the actor’s spatial trajectory.

Details

Authors
Organisations
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Robotics
  • Computer Systems
Original languageEnglish
Title of host publicationProceedings of the 9th International Conference on Agents and Artificial Intelligence (ICAART 2017)
PublisherSciTePress
Pages583-590
Number of pages8
ISBN (Electronic)978-989-758-220-2
Publication statusPublished - 2017
Publication categoryResearch
Peer-reviewedYes
EventICAART 2017-International Conference on Agents and Artificial Intelligence - Holiday Inn Porto Gaia, Porto, Portugal
Duration: 2017 Feb 242017 Feb 26

Conference

ConferenceICAART 2017-International Conference on Agents and Artificial Intelligence
CountryPortugal
CityPorto
Period2017/02/242017/02/26

Related research output

Gharaee, Z., 2018 Mar 15, Lund: Lund University Cognitive Science. 166 p.

Research output: ThesisDoctoral Thesis (compilation)

View all (1)

Related projects

Birger Johansson, Christian Balkenius, Magnus Johnsson, Rasmus Bååth, Stefan Winberg, Zahra Gharaee & Trond Arild Tjøstheim

2001/01/012021/12/31

Project: Research

View all (2)