Abstract
Embodied conversational agents that can sense and respond to multiple modalities of user communication, like speech, gesture, and facial expressions, create a better impression and facilitate communication [1,2]. Responding to a user’s gestures entails classifying the content and quality of each gesture, but classification performance is dependent on the selection of input sequence boundaries. Small changes in the boundaries of an input sequence can have a large effect on classifier output. Failing to correctly classify a user’s gestures may cause an agent to respond incorrectly, which can negatively impact the agent’s ability to communicate.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Cafaro, A., Vilhjálmsson, H.H., Bickmore, T., Heylen, D., Jóhannsdóttir, K.R., Valgarðsson, G.S.: First impressions: Users’ judgments of virtual agents’ personality and interpersonal attitude in first encounters. In: Nakano, Y., Neff, M., Paiva, A., Walker, M. (eds.) IVA 2012. LNCS, vol. 7502, pp. 67–80. Springer, Heidelberg (2012)
Huang, L., Morency, L.-P., Gratch, J.: Virtual rapport 2.0. In: Vilhjálmsson, H.H., Kopp, S., Marsella, S., Thórisson, K.R. (eds.) IVA 2011. LNCS, vol. 6895, pp. 68–79. Springer, Heidelberg (2011)
Zacharatos, H., Gatzoulis, C., Chrysanthou, Y., Aristidou, A.: Emotion recognition for exergames using laban movement analysis. In: Proceedings of Motion on Games, pp. 39–44. ACM (2013)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Bouchard, D. (2014). Improving Motion Classifier Robustness by Estimating Output Confidence. In: Bickmore, T., Marsella, S., Sidner, C. (eds) Intelligent Virtual Agents. IVA 2014. Lecture Notes in Computer Science(), vol 8637. Springer, Cham. https://doi.org/10.1007/978-3-319-09767-1_8
Download citation
DOI: https://doi.org/10.1007/978-3-319-09767-1_8
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-09766-4
Online ISBN: 978-3-319-09767-1
eBook Packages: Computer ScienceComputer Science (R0)