TY - GEN
T1 - Auto-generating virtual human behavior by understanding user contexts
AU - Kim, Hanseob
AU - Ali, Ghazanfar
AU - Kim, Seungwon
AU - Kim, Gerard J.
AU - Hwang, Jae In
N1 - Funding Information:
H. Kim and G. Ali contributed equally to this paper. This work has supported by the National Research Council of Science & Technology grant by the Korea Government (MSIP CRC-20-02-KIST).
Publisher Copyright:
© 2021 IEEE.
PY - 2021/3
Y1 - 2021/3
N2 - Virtual humans are most natural and effective when it can act out and animate verbal/gestural actions. One popular method to realize this is to infer the actions from predefined phrases. This research aims to provide a more flexible method to activate various behaviors straight from natural conversations. Our approach uses BERT as the backbone for natural language understanding and, on top of it, a jointly learned sentence classifier (SC) and entity classifier (EC). The SC classifies the input into conversation or action, and EC extracts the entities for the action. The pilot study has shown promising results with high perceived naturalness and positive experiences.
AB - Virtual humans are most natural and effective when it can act out and animate verbal/gestural actions. One popular method to realize this is to infer the actions from predefined phrases. This research aims to provide a more flexible method to activate various behaviors straight from natural conversations. Our approach uses BERT as the backbone for natural language understanding and, on top of it, a jointly learned sentence classifier (SC) and entity classifier (EC). The SC classifies the input into conversation or action, and EC extracts the entities for the action. The pilot study has shown promising results with high perceived naturalness and positive experiences.
KW - Computing methodologies
KW - Human-centered computing
KW - Interaction design process and methods
KW - Natural language processing
KW - Virtual reality
UR - http://www.scopus.com/inward/record.url?scp=85105966792&partnerID=8YFLogxK
U2 - 10.1109/VRW52623.2021.00178
DO - 10.1109/VRW52623.2021.00178
M3 - Conference contribution
AN - SCOPUS:85105966792
T3 - Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021
SP - 591
EP - 592
BT - Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021
Y2 - 27 March 2021 through 3 April 2021
ER -