Automatic text­to­gesture rule generation for embodied conversational agents

Title
Automatic text­to­gesture rule generation for embodied conversational agents
Authors
황재인이명호가잔파 알리
Issue Date
2020-07
Publisher
Computer Animation and Virtual Worlds
Citation
VOL 31, NO 4-5, e1944
Abstract
Interactions with embodied conversational agents can be enhanced using human­like co­speech gestures. Traditionally, rule­based co­speech gesture mapping has been utilized for this purpose. However, the creation of this mapping is laborious and often requires human experts. Moreover, human­created mapping tends to be limited, therefore prone to generate repeated gestures. In this article, we present an approach to automate the generation of rule­based co­speech gesture mapping from publicly available large video data set without the intervention of human experts. At run­time, word embedding is utilized for rule searching to get the semantic­aware, meaningful, and accurate rule. The evaluation indicated that our method achieved comparable performance with the manual map generated by human experts, with a more variety of gestures activated. Moreover, synergy effects were observed in users' perception of generated co­speech gestures when combined with the manual map.
URI
http://pubs.kist.re.kr/handle/201004/72239
ISSN
1546-4261
Appears in Collections:
KIST Publication > Article
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

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

BROWSE