Rolf P. Würtz

Ruhr University Bochum

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Christoph von der Malsburg

Frankfurt Institute for Advanced Studies

Kirstie L. Bellman

University of Bologna

Manuel Günther

University of Zurich

Dagmar Wieczorek

Düsseldorf University Hospital

Thomas Walther

TU Dresden

Bernhard Horsthemke

University of Duisburg-Essen

Martín Reincke

Ludwig-Maximilians-Universität München

Robert P. Kosilek

Ludwig-Maximilians-Universität München

Andreas Tewes

Laser Zentrum Hannover

Marco K. Müller

Brandenburg University of Technology Cottbus-Senftenberg

Cooperative Institutions

Ruhr University Bochum

Ludwig-Maximilians-Universität München

LMU Klinikum

Max Planck Institute of Psychiatry

München Klinik

University of Bonn

Essen University Hospital

Max Planck Society

Erasmus University Rotterdam

Idiap Research Institute

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Two Kinds of Statistics for Better Face Recognition

AIP conference proceedings (2010)

Manuel Günther Marco K. Müller Rolf P. Würtz Theodore E. Simos George Psihoyios

We briefly review the base techniques of elastic graph matching [1] and elastic bunch graph matching [2], which provide a method for face detection, matching, comparison, and identity decision. We then present a method that combines the advantages of Gabor‐labeled graphs with maximum likelihood decision making. The improvements over pure bunch graph matching have been studied, and the method has been successfully applied to large face databases like the one for the Face Recognition Grand Challenge [3]. Finally, we describe a system based on rank order statistics that can learn invariances in a minimally supervised way from a set of examples of individual faces in several situations like different head pose or illumination. Recognition rates are improved significantly without explicit modeling of the image formation process [4].

Rank (graph theory)

10.1063/1.3498288

Cite

Citations (2)

Face Recognition, Neurophysiology, and Neural Technology

Rolf P. Würtz

The recognition of other individuals is a major prerequisite for human social interaction and hence a rather important brain function. The most prominent cue for that recognition is the face. The capability to recognize persons from their faces is part of a spectrum of related skills, which include face segmentation, i.e., finding faces in a scene or image, the estimation of the pose, the direction of gaze, and the person’s emotional state. This article focuses on recognition of identity, a more detailed treatment of the other aspects can be found in [Face Recognition, Psychology, and Connectionism].

Source

Cite

Citations (0)

Classification and Visualization Based on Derived Image Features: Application to Genetic Syndromes

PLoS ONE (2014)

Brunilda Balliu Rolf P. Würtz Bernhard Horsthemke Dagmar Wieczorek Stefan Böhringer

Data transformations prior to analysis may be beneficial in classification tasks. In this article we investigate a set of such transformations on 2D graph-data derived from facial images and their effect on classification accuracy in a high-dimensional setting. These transformations are low-variance in the sense that each involves only a fixed small number of input features. We show that classification accuracy can be improved when penalized regression techniques are employed, as compared to a principal component analysis (PCA) pre-processing step. In our data example classification accuracy improves from 47% to 62% when switching from PCA to penalized regression. A second goal is to visualize the resulting classifiers. We develop importance plots highlighting the influence of coordinates in the original 2D space. Features used for classification are mapped to coordinates in the original images and combined into an importance measure for each pixel. These plots assist in assessing plausibility of classifiers, interpretation of classifiers, and determination of the relative importance of different features.

Data set

10.1371/journal.pone.0109033

Cite

Citations (9)

EXTRACTION AND MATCHING OF SYMBOLIC CONTOUR GRAPHS

International Journal of Pattern Recognition and Artificial Intelligence (2003)

Tino Lourens Rolf P. Würtz

We describe an object recognition system based on symbolic contour graphs. The image to be analyzed is transformed into a graph with object corners as vertices and connecting contours as edges. Image corners are determined using a robust multiscale corner detector. Edges are constructed by line-following between corners based on evidence from the multiscale Gabor wavelet transform. Model matching is done by finding subgraph isomorphisms in the image graph. The complexity of the algorithm is reduced by labeling vertices and edges, whereby the choice of labels also makes the recognition system invariant under translation, rotation and scaling. We provide experimental evidence and theoretical arguments that the matching complexity is below O(#V 3 ), and show that the system is competitive with other graph-based matching systems.

Gabor wavelet

10.1142/s0218001403002848

Cite

Citations (9)

Face authentication test on the BANCA database

Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004. (2004)

Kieron Messer Josef Kittler Mohammad Taghi Sadeghi M. Hamouz A. Kostin

This work details the results of a face authentication test (FAT2004) (http://www.ee.surrey.ac.uk/banca/icpr2004) held in conjunction with the 17th International Conference on Pattern Recognition. The contest was held on the publicly available BANCA database (http://www.ee.surrey.ac.uk/banca) according to a defined protocol (E. Bailly-Bailliere et al., June 2003). The competition also had a sequestered part in which institutions had to submit their algorithms for independent testing. 13 different verification algorithms from 10 institutions submitted results. Also, a standard set of face recognition software packages from the Internet (http://www.cs.colostate.edu/evalfacerec) were used to provide a baseline performance measure.

CONTEST

10.1109/icpr.2004.1333826

Cite

Citations (64)

Emergent Graphs with PCA-features for Improved Face Recognition

AIP conference proceedings (2006)

A. Heinrichs Marco K. Müller Andreas Tewes Rolf P. Würtz

Built on the principles of “Learning from Nature” and “Self‐organization” Elastic Bunch Graph Matching for face recognition is a defining example for Organic Computing methodology. Here, we follow these principles further to advance the method in two respects. First, the requirement for manual annotation of landmarks is reduced to one single face, from which a self‐organizing selection process gradually builds up the bunches by adding the most similar face to the bunch graph and then recalculating the matching. Second, the resulting bunches are replaced by the principal components of the nodes of all persons in the database. The similarity function is restricted to a suitable subset of these components. The additional self‐organizing processes lead to improved precision of landmark localization and recognition rates. Altogether, an improved data structure for face storage has emerged from the simple presentation of examples in a minimally supervised way.

Landmark

10.1063/1.2361226

Cite

Citations (5)

Organic Computing - Design of Self-Organizing Systems

Kirstie L. Bellman Andreas Herkersdorf Mike Hinchey Mike Hinchey Christian Müller-Schloer

This report documents the program and the outcomes of Dagstuhl Seminar 11181 “Organic Computing ‐ Design of Self-Organizing Systems”. Seminar 1.‐6. May, 2011 ‐ www.dagstuhl.de/11181 1998 ACM Subject Classification F.1.1 Models of computation.

Source

Cite

Citations (0)

4th Workshop on Dynamic Perception.

Künstliche Intell. (2003)

Rolf P. Würtz

Source

Cite

Citations (0)

Using Growing Neural Gas Networks to Represent Visual Object Knowledge

Guillermo S. Donatti Rolf P. Würtz

We present a so-called Neural Map, a novel memory framework for visual object recognition and categorization systems. The properties of its computational theory include self-organization and intelligent matching of the image features that are used to build their object models. Its performance for representing the visual object knowledge comprised by these models and for recognizing unknown objects is measured using three different types of image features, which extract different granularity of information from object views of the ETH-80 image set. The obtained experimental results slightly outperform previous ones using PCA-based methods on the same image set, and they suggest that the medium-sized image features maximize the object models' informativeness and distinctiveness.

Optimal distinctiveness theory

Granularity

10.1109/ictai.2009.81

Cite

Citations (6)

Gossiping Nets

Artificial Intelligence (2000)

Rolf P. Würtz

10.1016/s0004-3702(00)00015-1

Cite

Citations (0)