English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Conference Paper

View-based recognition of faces in man and machine: Re-visiting Inter-Extra-Ortho

MPS-Authors
/persons/resource/persons84298

Wallraven,  C
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons84420

Schwaninger,  A
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons83839

Bülthoff,  HH
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

Fulltext (public)
There are no public fulltexts stored in PuRe
Supplementary Material (public)
There is no public supplementary material available
Citation

Wallraven, C., Schwaninger, A., Schumacher, S., & Bülthoff, H. (2002). View-based recognition of faces in man and machine: Re-visiting Inter-Extra-Ortho. In Biologically Motivated Computer Vision: Second International Workshop, BMCV 2002 Tübingen, Germany, November 22–24, 2002 (pp. 651-660). Berlin, Germany: Springer.


Cite as: http://hdl.handle.net/11858/00-001M-0000-0013-DE5C-A
Abstract
For humans, faces are highly overlearned stimuli, which are encountered in everyday life in all kinds of poses and views. Using psychophysics we investigated the effects of viewpoint on human face recognition. The experimental paradigm is modeled after the inter-extra-ortho experiment using unfamiliar objects by Bülthoff and Edelman [5]. Our results show a strong viewpoint effect for face recognition, which replicates the earlier findings and provides important insights into the biological plausibility of view-based recognition approaches (alignment of a 3D model, linear combination of 2D views and view-interpolation). We then compared human recognition performance to a novel computational view-based approach [29] and discuss improvements of view-based algorithms using local part-based information.