Face perception is critical for normal social functioning, and is mediated by a cortical network of regions in the ventral visual stream. Comparative analysis between present deep neural network architectures for biometrics and neural architectures in the human brain is necessary for developing artificial systems with human abilities. Neuroimaging research has advanced our understanding regarding the functional architecture of the human ventral face network. Here, we describe recent neuroimaging findings in three domains: (1) the macro- and microscopic anatomical features of the ventral face network in the human brain, (2) the characteristics of white matter connections, and (3) the basic computations performed by population receptive fields within face-selective regions composing this network. Then, we consider how empirical findings can inform the development of accurate computational deep neural networks for face recognition as well as shed light on computational benefits of specific neural implementational features.
|Original language||English (US)|
|Title of host publication||Advances in Computer Vision and Pattern Recognition|
|Number of pages||29|
|State||Published - 2017|
|Name||Advances in Computer Vision and Pattern Recognition|
Bibliographical noteFunding Information:
Acknowledgements We thank Jesse Gomez for comments and edits of an earlier draft. This research has been funded by the following grants: NSF GRFP DGE-114747, 1RO1EY02231801-A1, and 1R01EY02391501-A1. Portions of this chapter were previously included within a paper published in the Annual Reviews of Vision Science, 2017 (http://www.annualreviews.org/doi/10. 1146/annurev-vision-102016-061214).
© Springer International Publishing AG 2017.