We study the problem of expression analysis for a group of people. Automatic facial expression analysis has seen much research in recent times. However, little attention has been given to the estimation of the overall expression theme conveyed by an image of a group of people. Specically, this work focuses on formulating a framework for happiness intensity estimation for groups based on social context information. The main contributions of this paper are: a) dening automatic frameworks for group expressions; b) social features, which compute weights on expression intensities; c) an automatic face occlusion intensity detection method; and d) an `in the wild' labelled database containing images having multiple subjects from different scenarios. The experiments show that the global and local contexts provide useful information for theme expression analysis, with results similar to human perception results.
Few images from the HAPPEI database.
Current FER databases are individual based. Flickr search was performed for collecting data containing various scenarios depicting various social events. The database images have been labelled for the mood of the group members, pose, face clearity and group's mood perception. We call the database: HAPpy PEople Images (HAPPEI). Figure above displays a collage of images in HAPPEI.
Download HAPPEI database
A. Dhall, R. Goecke and T. Gedeon, Automatic Group Happiness Intensity Analysis, IEEE Transaction on Affective Computing 2015[PDF].
A. Dhall, J. Joshi, I. Radwan and R. Goecke, Finding Happiest Moments in a Social Context.Proceedings of the 11th Asian Conference on Computer Vision ACCV2012, Daejeon, Korea [PDF] [Supplementary Material].
A. Dhall and R. Goecke, Group Expression Intensity Estimation in Videos via Gaussian Processes.Proceedings of the International Conference on Pattern Recognition ICPR2012, Tsukuba, Japan, 11-15 Nov 2012.