Paper Info
Title
Towards Integrative Multimedia Understanding for Cognitive Robotics Based on Mental Image Directed Semantic Theory
Abstract
This paper describes a methodology to provide robots with capability of integrative multimedia understanding centered in natural language understanding (NLU) and its recent application to cross-media operation between linguistic and pictorial representations such as town maps and animations of robotic action. The core of the methodology is based on the mental image directed semantic theory (MIDST). MIDST is intended to simulate omnisensory mental imagery in human and to afford a knowledge representation system in order for integrative management of knowledge subjective to cognitive mechanisms of intelligent entities such as humans and robots based on a mental image model and its representation in the formal language Lmd. At present stage, the developed NLU system is still experimental with a small vocabulary (about 100 words) but can perform human-like cross-media operation as well as expected.
Year
DOI
Venue
2017
10.1109/CYBConf.2017.7985818
2017 3rd IEEE International Conference on Cybernetics (CYBCONF)
Keywords
Field
DocType
integrative multimedia understanding,cognitive robotics,mental image directed semantic theory,cross-media operation,semantic awareness,natural language understanding,NLU,linguistic representations,pictorial representations,MIDST,omnisensory mental imagery,knowledge representation system,knowledge integrative management,intelligent entities cognitive mechanisms,formal language Lmd,vocabulary
Cognitive robotics,Knowledge representation and reasoning,Formal language,Computer science,Mental image,Natural language understanding,Cognition,Robot,Vocabulary,Multimedia
Conference
ISBN
Citations 
PageRank 
978-1-5386-2202-5
0
0.34
References 
Authors
3
2
Name
Order
Citations
PageRank
Rojanee Khummongkol100.34
Masao Yokota220.71