Initial perceptions of a casual game to crowdsource facial expressions in the wild

Publisher:
Society for the Advancement of the Study of Digital Games
Publication Type:
Conference Proceeding
Citation:
Proceedings of the 9th International Conference on the Foundations of Digital Games, 2014, pp. 1 - 4
Issue Date:
2014-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnailfdg2014_wip_19.pdfAccepted Manuscript version370.59 kB
Adobe PDF
The performance of affective computing systems often depend on the quality of the image databases they are trained on. However, creating good quality training databases is a laborious activity. In this paper, we evaluate BeFaced, a tile matching casual tablet game that enables massive crowdsourcing of facial expressions for the purpose of advancing facial expression analysis. The core aspect of BeFaced is game quality, as increased enjoyment and engagement translates to an increased quantity of varied facial expressions obtained. Hence a pilot user study was performed on 18 university students whereby observational and interview data were obtained during playtests. We found that most users enjoyed the game and were intrigued by the novelty in interacting with the facial expression gameplay mechanic, but also uncovered problems with feedback provision and the dynamic difficulty adjustment mechanism. These findings hence provide invaluable insights for the other researchers/ practitioners working on similar crowdsourcing games with a purpose, as well as for the development of BeFaced.
Please use this identifier to cite or link to this item: