Please use this identifier to cite or link to this item:https://hdl.handle.net/20.500.12259/38678
Type of publication: Straipsnis recenzuojamoje užsienio tarptautinės konferencijos medžiagoje / Article in peer-reviewed foreign international conference proceedings (P1d)
Field of Science: Informatika / Informatics (N009)
Author(s): Vidugirienė, Aušra;Vaškevičius, Egidijus;Kaminskas, Vytautas
Title: Two predictive models for identification of human responses to virtual 3D face stimuli
Is part of: ECT-2014 : Electrical and control technologies : proceedings of the 9th international conference on electrical and control technologies, May 8-9, 2014, Kaunas, Lithuania. Kaunas : Technologija, 2014, no. 9
Extent: p. 78-83
Date: 2014
Keywords: Trimatis veidas;Žmogaus emocijos;Įėjimo-išėjimo modelis;Parametrų įvertinimas;Prognozavimas;3D face;Human emotions;Input-output model;Parameter estimation;Prediction
Abstract: This paper introduces a comparison of two one step predictive models that were used to model the relationship between human emotional signals – excitement and frustration – and three virtual 3D face features – distance between eyes, nose width, and chin width. Two volunteers’ data werezas used for the comparison. Two input-output model building methods are proposed that allow building stable models with the least output prediction error. Validation results of the models demonstrate relatively high prediction accuracy of human reactions and both models produce similar prediction accuracy
Internet: http://www.bpti.lt/uploads/Proceedings%20ECT-2014.pdf
Affiliation(s): Informatikos fakultetas
Sistemų analizės katedra
Vytauto Didžiojo universitetas
Appears in Collections:Universiteto mokslo publikacijos / University Research Publications

Files in This Item:
marc.xml8.9 kBXMLView/Open

MARC21 XML metadata

Show full item record
Export via OAI-PMH Interface in XML Formats
Export to Other Non-XML Formats

Page view(s)

142
checked on Mar 5, 2020

Download(s)

12
checked on Mar 5, 2020

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.