Proc. Int'l Conf. on Artificial Neural Networks, ICANN'95, Paris, Oct. 9-13, eds. F. Fogelman-Soulié, J. C. Rault, P. Gallinari, and G. Dreyfus, publ. EC2 & Cie, pp. 347-352 (1995-10-09) (bibtex)

Reprinted: Proc. Symposium über Biologische Informationsverarbeitung und Neuronale Netze, SINN'95, eds. A. Wismüller, and D.R. Dersch, series Beiträge zur wissenschaftlichen Diskussion, publ. Hanns-Seidel-Stiftung, München, pp. 63-68 (1996) (bibtex)

Reprinted: Proc. US-EC Workshop on Neuroinformatics '95, Washington DC, Sep 20, special issue of NeuroImage, 4(3):S14-S18 (1996-09-20) (bibtex)


Recognizing faces by dynamic link matching.

Laurenz Wiskott and Christoph von der Malsburg


Abstract: Dynamic Link Matching is a neural dynamics for translation invariant object recognition that is robust against distortion. We here demonstrate human face recognition against a gallery of 112 neutral frontal view faces. Probe images are distorted due to rotation in depth and changing facial expression. Probe images and gallery models are represented by layers of neurons interpreted as labeled graphs. Nodes are labeled by local features based on the Gabor wavelet transform. Probe images are matched to the gallery of face models by Dynamic Link Matching. Concurrently with the matching process a simple winner-take-all mechanism identifies the correct model. A dynamic window of attention restricts the match to the part of the image occupied by the face.

Keywords: neural networks, dynamic link matching, face recognition, translation invariance, window of attention


Relevant Project:


December 14, 1998, Laurenz Wiskott, http://www.neuroinformatik.ruhr-uni-bochum.de/PEOPLE/wiskott/