ARTICLE
Hung_IEEETRANS.ONMULTIMEDIA_2010/IDIAP
Estimating Cohesion in Small Groups Using Audio-Visual Nonverbal Behavior
Hung, Hayley
Gatica-Perez, Daniel
cohesion
group interaction
social psychology
https://publications.idiap.ch/index.php/publications/showcite/Hung_Idiap-RR-12-2010
Related documents
IEEE Trans. on Multimedia, Special Issue on Multimodal Affective Interaction
12
6
563 - 575
2010
Cohesiveness in teams is an essential part of ensuring
the smooth running of task-oriented groups. Research
in social psychology and management has shown that good
cohesion in groups can be correlated with team effectiveness
or productivity so automatically estimating group cohesion for
team training can be a useful tool. This paper addresses the
problem of analyzing group behavior within the context of
cohesion. 4 hours of audio-visual group meeting data was used
for collecting annotations on the cohesiveness of 4-participant
teams. We propose a series of audio and video features, which
are inspired by findings in the social sciences literature. Our
study is validated on as set of 61 2-minute meeting segments
which showed high agreement amongst human annotators who
were asked to identify meetings which have high or low cohesion.
REPORT
Hung_Idiap-RR-12-2010/IDIAP
Estimating Cohesion in Small Groups using Audio-Visual Nonverbal Behavior
Hung, Hayley
Gatica-Perez, Daniel
Idiap-RR-12-2010
2010
Idiap
June 2010
Cohesiveness in teams is an essential part of ensuring
the smooth running of task-oriented groups. Research
in social psychology and management has shown that good
cohesion in groups can be correlated with team effectiveness
or productivity so automatically estimating group cohesion for
team training can be a useful tool. This paper addresses the
problem of analyzing group behavior within the context of
cohesion. 4 hours of audio-visual group meeting data was used
for collecting annotations on the cohesiveness of 4-participant
teams. We propose a series of audio and video features, which
are inspired by findings in the social sciences literature. Our
study is validated on as set of 61 2-minute meeting segments
which showed high agreement amongst human annotators who
were asked to identify meetings which have high or low cohesion.