Avatars are pervasive in video games and virtual worlds. The automatic generation of these avatar promises to reduce player effort and provide system-defined mappings between “real” (physical) player characteristics and virtual identities. We present an avatar generation system called Exigent; given a photograph of a human face, Exigent creates an avatar. Exigent leverages two recent computer vision packages and extends them with a custom facial analysis and generation sub-module. The generated avatars incorporate the user’s gender, face shape, hair color, facial expression, eye shape, and more. This system is different from most avatar creators in its consideration of finer-grained details such as curvature of the face, distance between eyes, size of features, etc.
[1]
Fernando De la Torre,et al.
Supervised Descent Method and Its Applications to Face Alignment
,
2013,
2013 IEEE Conference on Computer Vision and Pattern Recognition.
[2]
D. Fox Harrell.
Phantasmal Media: An Approach to Imagination, Computation, and Expression
,
2013
.
[3]
Deva Ramanan,et al.
Face detection, pose estimation, and landmark localization in the wild
,
2012,
2012 IEEE Conference on Computer Vision and Pattern Recognition.
[4]
Falko Kuester,et al.
Automatic creation of three-dimensional avatars
,
2003,
IS&T/SPIE Electronic Imaging.
[5]
J. Goguen.
An introduction to algebraic semiotics, with application to user interface design
,
1999
.