Hierarchical self-organizing maps system for action classification

Zahra Gharaee, Peter Gärdenfors, Magnus Johnsson

Forskningsoutput: KonferensbidragArbetsdokument (paper)Peer review

7 Citeringar (Scopus)

Sammanfattning

We present a novel action recognition system that is able to learn how to recognize and classify actions. Our system employs a three-layered neural network hierarchy consisting of two self-organizing maps together with a supervised neural network for labelling the actions. The system is equipped with a module that preprocesses the 3D input data before the first layer, and a module that transforms the activity elicited over time in the first layer SOM into an ordered vector representation before the second layer, thus achieving a time invariant representation. We have evaluated our system in an experiment consisting of ten different actions selected from a publicly available data set with encouraging result.

OriginalspråkEngelska
Sidor583-590
Antal sidor8
DOI
StatusPublicerad - 2017
Externt publiceradJa
Evenemang9th International Conference on Agents and Artificial Intelligence, ICAART 2017 - Porto, Portugal
Varaktighet: 2017-feb.-242017-feb.-26

Konferens

Konferens9th International Conference on Agents and Artificial Intelligence, ICAART 2017
Land/TerritoriumPortugal
OrtPorto
Period17-02-2417-02-26

Fingeravtryck

Fördjupa i forskningsämnen för ”Hierarchical self-organizing maps system for action classification”. Tillsammans bildar de ett unikt fingeravtryck.

Citera det här