KIT | KIT-Bibliothek | Impressum
Open Access Logo
§
Volltext
URN: urn:nbn:de:swb:90-447747

Multimodal Computational Attention for Scene Understanding

Schauerte, Boris

Abstract:
Robotic systems have limited computational capacities. Hence, computational attention models are important to focus on specific stimuli and allow for complex cognitive processing. For this purpose, we developed auditory and visual attention models that enable robotic platforms to efficiently explore and analyze natural scenes. To allow for attention guidance in human-robot interaction, we use machine learning to integrate the influence of verbal and non-verbal social signals into our models.


Zugehörige Institution(en) am KIT Institut für Anthropomatik und Robotik (IAR)
Publikationstyp Hochschulschrift
Jahr 2014
Sprache Englisch
Identifikator KITopen ID: 1000044774
Verlag Karlsruhe
Abschlussart Dissertation
Fakultät Fakultät für Informatik (INFORMATIK)
Institut Institut für Anthropomatik und Robotik (IAR)
Prüfungsdaten 13.06.2014
Referent/Betreuer Prof. R. Stiefelhagen
Schlagworte Attention, Visual Saliency, Auditory Saliency, Scene Exploration, Human-Robot Interaction
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft KITopen Landing Page