Heiko Wersing, Stephan Kirstein, Bernd Schneiders, Ute Bauer-wersing, Edgar Körner
Abstract. We present a modular architecture for recognition and localization of objects in a scene that is motivated from coupling the ventral (“what”) and dorsal (“where”) pathways of human...
Approaches and Challenges for Cognitive Vision Systems (2009)
Abstract. A cognitive visual system is generally intended to work robustly under varying environmental conditions, adapt to a broad range of unforeseen changes, and even exhibit prospective behavior...
Unsupervised Extraction of Design Components for a 3D parts-based Representation (2009)
Zdravko Bozakov, Lars Graening, Stephan Hasler, Heiko Wersing, Stefan Menzel
Abstract—During CAD development and any kind of design optimisation over years a huge amount of geometries accumulate in a design department. To organize and structure these designs with respect to...
A Vector Quantization Approach for Life-Long Learning of Categories (2009)
Stephan Kirstein, Heiko Wersing, Horst-michael Gross, Edgar Körner
Abstract. We present a category learning vector quantization (cLVQ) approach for incremental and life-long learning of multiple visual categories where we focus on approaching the...
An Integrated System for Incremental Learning of Multiple Visual Categories (2009)
Stephan Kirstein, Heiko Wersing, Horst-michael Gross, Edgar Körner
Abstract. We present a biologically inspired vision system able to incrementally learn multiple visual categories by interactively presenting several hand-held objects. The overall system is composed...
Online Figure-Ground Segmentation with Adaptive Metrics in Generalized LVQ (2009)
Er Denecke, Heiko Wersing, Jochen J. Steil, Edgar Körner
We address the problem of fast figure-ground segmentation of single objects from cluttered backgrounds to improve object learning and recognition. For the segmentation, we use an initial foreground...
Robust object segmentation by adaptive metrics in Generalized LVQ (2009)
Er Denecke, Heiko Wersing, Jochen J. Steil, Edgar Körner
Abstract. We investigate the effect of several adaptive metrics in the context of figure-ground segregation, using Generalized LVQ to train a classifier for image regions. Extending the Euclidean...
Towards Incremental Hierarchical Behavior Generation for Humanoids (2008)
Christian Goerick, Bram Bolder, Herbert Janßen, Michael Gienger, Hisashi Sugiura, Mark Dunn, ...
Abstract — The contribution of this paper is twofold. First, we present a new conceptual framework for modeling incremental hierarchical behavior control systems for humanoids. The biological...
Evolution of Hierarchical Features for Visual Object Recognition (2008)
Georg Schneider, Heiko Wersing, Bernhard Sendhoff, Edgar Körner
A central task in visual object recognition applications is the choice of a proper feature space. Inspired from the human neural vision system we suggest hierarchical features for 3D object...
There is an ongoing debate over the capabilities of hierarchical neural feed-forward architectures for performing real-world invariant object recognition. Although a variety of hierarchical models...
Sebastian Weng, Heiko Wersing, Jochen J. Steil, Helge Ritter
Abstract — We present a hybrid learning method bridging the fields of recurrent neural networks, unsupervised Hebbian learning, vector quantization, and supervised learning to implement a...
Combining Reconstruction and Discrimination with Class-specific Sparse Coding (2008)
Stephan Hasler, Heiko Wersing, Edgar Körner
Sparse coding is an important approach for the unsupervised learning of sensory features. In this contribution we present two new methods which extend the traditional sparse coding approach with...
A Biologically Motivated Visual Memory Architecture for Online Learning of Objects Abstract (2008)
Stephan Kirstein, Heiko Wersing, Edgar Körner
We present a biologically motivated architecture for object recognition that is based on a hierarchical feature detection model in combination with a memory architecture that implements short-term...
Ra Mark, Bernhard Sendhoff, Heiko Wersing
We describe an adaptive decision making architecture which is applied to competitive games. The task is to learn online a model of the current opponent strategy which is used to predict the next...
Word Recognition with a Hierarchical Neural Network (2008)
Xavier Domont, Martin Heckmann, Heiko Wersing, Frank Joublin, Stefan Menzel, Bernhard Sendhoff, ...
In this paper we propose a feedforward neural network for syllable recognition. The core of the recognition system is based on a hierarchical architecture initially developed for visual object...
Christian Goerick, Inna Mikhailova, Heiko Wersing, Stephan Kirstein
Biologically motivated visual behaviors for
A Comparison of Features in Parts-Based Object Recognition Hierarchies (2008)
Stephan Hasler, Heiko Wersing, Edgar Körner
Abstract. Parts-based recognition has been suggested for generalizing from few training views in categorization scenarios. In this paper we present the results of a comparative investigation of...
Stephan Kirstein, Heiko Wersing, Edgar Körner
We present a biologically motivated architecture for object recognition that is based on a hierarchical feature detection model in combination with a memory architecture that implements short-term...
Evolutionary feature design for object recognition with hierarchical networks (2007)
Georg Schneider, Heiko Wersing, Bernhard Sendhoff, Edgar Körner
A major problem in designing neural vision models is the large dimensionality of the search space for defining the needed networks. By using hierarchical vision models inspired by biology we narrow...
There is an ongoing debate over the capabilities of hierarchical neural feed-forward architectures for performing real-world invariant object recognition. Although a variety of hierarchical models...
Using Maximal Recurrence in Linear Threshold Competitive Layer Networks (2007)
Abstract. We demonstrate the application of recent theoretical results on the stability of linear threshold (LT) networks to the competitive layer model architecture (CLM). LT networks can be...
BACKTRACKING DETERMINISTIC ANNEALING FOR CONSTRAINT SATISFACTION PROBLEMS (2007)
We present a new deterministic annealing approach to the solution of quadratic constraint satisfaction problems with complex interlocking constraints, such as exemplified in polyomino tiling puzzles....
Tim W. Nattkemper, Heiko Wersing, Helge Ritter, Walter Schubert
We present the application of a recurrent neural network feature binding model to the segmentation of uorescence micrographs, images showing uorescent cells in tonsil tissue. Image primitives,...
Exploiting Ensemble Diversity For (2007)
Automatic Feature Extraction, Gavin Brown, Xin Yao, Jeremy Wyatt, Heiko Wersing, Bernhard Sendhoff
We present an automatic method, based on a neural network ensemble, for extracting multiple, diverse and complementary sets of useful classification features from highdimensional data. We demonstrate...
Online Learning of Objects and Faces in an Integrated Biologically Motivated Architecture (2007)
Wersing, Heiko, Kirstein, Stephan, Goetting, Michael, Brandl, Holger, Dunn, Mark, Mikhailova, Inna, ...
We present a biologically motivated integrated vision system that is capable of online learning of several objects and faces in a unified representation. The training is unconstrained in the sense...
Online Learning of Objects in a Biologically Motivated Visual Architecture (2007)
Heiko Wersing, Stephan Kirstein, Michael Götting, Holger Br, Mark Dunn, Inna Mikhailova, ...
We present a biologically motivated architecture for object recognition that is capable of online learning of several objects based on interaction with a human teacher. The system combines biological...
A Hierarchical Model for Syllable Recognition (2007)
Xavier Domont, Martin Heckmann, Heiko Wersing, Frank Joublin, Christian Goerick
Inspired by recent findings on the similarities between the primary auditory and visual cortex we propose a neural network for speech recognition based on a hierarchical feedforward architecture for...
Online learning of objects and faces in an integrated biologically motivated architecture (2007)
Heiko Wersing, Stephan Kirstein, Michael Götting, Holger Br, Inna Mikhailova, Christian Goerick, ...
Abstract. We present a biologically motivated integrated vision system that is capable of online learning of several objects and faces in a unified representation. The training is unconstrained in...
A biologically motivated system for unconstrained online learning of visual objects (2006)
Heiko Wersing, Stephan Kirstein, Michael Götting, Holger Br, Inna Mikhailova, Christian Goerick, ...
of visual objects
Recent trends in online learning for cognitive robotics (2006)
Jochen J. Steil, Heiko Wersing
Abstract. We present a review of recent trends in cognitive robotics that deal with online learning approaches to the acquisition of knowledge, control strategies and behaviors of a cognitive robot...
E.: Class-specific Sparse Coding for Learning of Object Representations (2005)
Stephan Hasler, Heiko Wersing, Edgar Körner
Abstract. We present two new methods which extend the traditional sparse coding approach with supervised components. The goal of these extensions is to increase the suitability of the learned...
Peripersonal space and object recognition for humanoids (2005)
Christian Goerick, Heiko Wersing, Inna Mikhailova, Mark Dunn
Abstract — This work is concerned with a framework for visual object recognition in real world tasks. Our approach is motivated by biological findings of the representation of space around the...
Online learning for object recognition with a hierarchical visual cortex model (2005)
Stephan Kirstein, Heiko Wersing, Edgar Körner
Abstract. We present an architecture for the online learning of object representations based on a visual cortex hierarchy developed earlier. We use the output of a topographical feature hierarchy to...
Rapid online learning of objects in a biologically motivated recognition architecture (2005)
Stephan Kirstein, Heiko Wersing, Edgar Körner
Abstract. We present an approach for the supervised online learning of object representations based on a biologically motivated architecture of visual processing. We use the output of a recently...
Evolutionary optimization of an hierarchical object recognition model (2005)
Georg Schneider, Heiko Wersing, Bernhard Sendhoff, Edgar Körner
Abstract — A major problem in designing artificial neural networks is the proper choice of the network architecture. Especially for vision networks classifying 3D objects this problem is very...
A computational feature binding model of human texture perception (2004)
Jörg Ontrup, Helge Ritter, Heiko Wersing
We present a computational model for human texture perception which assigns functional principles to the Gestalt laws of similarity and proximity. Motivated by early vision mechanisms, in a first...
Transformation-invariant representation and NMF (2004)
Julian Eggert, Heiko Wersing, Edgar Körner
Abstract — Non-negative matrix factorization (NMF) is a method for the decomposition of multivariate data into strictly positive activations and basis vectors. Here, instead of using unstructured...
Coupling of evolution and learning to optimize a hierarchical object recognition model (2004)
Georg Schneider, Heiko Wersing, Bernhard Sendhoff, Edgar Körner
Abstract. A key problem in designing artificial neural networks for visual object recognition tasks is the proper choice of the network architecture. Evolutionary optimization methods can help to...
Sparse coding with invariance constraints (2003)
Heiko Wersing, Julian Eggert, Edgar Körner
Abstract. We suggest a new approach to optimize the learning of sparse features under the constraints of explicit transformation symmetries imposed on the set of feature vectors. Given a set of basis...
Sparse coding with invariance constraints (2003)
Heiko Wersing, Julian Eggert, Edgar Körner
Abstract. We suggest a new approach to optimize the learning of sparse features under the constraints of explicit transformation symmetries imposed on the set of feature vectors. Given a set of basis...
Learning Optimized Features for Hierarchical Models of Invariant Object Recognition (2003)
There is an ongoing debate over the capabilities of hierarchical neural feed-forward architectures for performing real-world invariant object recognition. Although a variety of hierarchical models...
Unsupervised learning of combination features for hierarchical recognition models (2002)
Abstract. We propose a cortically inspired hierarchical feedforward model for recognition and investigate a new method for learning optimal combination-coding cells in intermediate stages of the...
A neural network architecture for automatic segmentation of fluorescence micrographs (2002)
Tim W. Nattkemper, Heiko Wersing, Walter Schubert, Helge Ritter
Learning lateral interactions for feature binding and sensory segmentation (2001)
We present a new approach to the supervised learning of lateral interactions for the competitive layer model (CLM) dynamic feature binding architecture. The method is based on consistency conditions,...
A competitive layer model for feature binding and sensory segmentation (2001)
Heiko Wersing, Jochen J. Steil, Helge Ritter
We present a recurrent neural network for feature binding and sensory segmentation, the competitive layer model (CLM). The CLM uses topographically structured competitive and cooperative interactions...
Heiko Wersing, Wolf-jurgen Beyn, Helge Ritter
We establish two conditions which ensure the non-divergence of additive recurrent networks with unsaturating piecewise linear transfer functions, also called linear threshold or semilinear transfer...
Learning lateral interactions for feature binding and sensory segmentation (2001)
We present a new approach to the supervised learning of lateral interactions for the competitive layer model (CLM) dynamic feature binding architecture. The method is based on consistency conditions,...
Heiko Wersing, Wolf-jurgen Beyn, Helge Ritter
to appear in Neural Computation We establish two conditions which ensure the non-divergence of additive recurrent networks with unsaturating piecewise linear transfer functions, also called linear...
A competitive layer model for feature binding and sensory segmentation (2001)
Heiko Wersing, Jochen J. Steil, Helge Ritter
We present a recurrent neural network for feature binding and sensory segmentation, the competitive layer model (CLM). The CLM uses topographically structured competitive and cooperative interactions...
Learning lateral interactions for feature binding and sensory segmentation (2001)
We present a new approach to the supervised learning of lateral interactions for the competitive layer model (CLM) dynamic feature binding architecture. The method is based on consistency conditions,...
Learning lateral interactions for feature binding and sensory segmentation (2001)
We present a new approach to the supervised learning of lateral interactions for the competitive layer model (CLM) dynamic feature binding architecture. The method is based on consistency conditions,...
Heiko Wersing, Wolf-jürgen Beyn, Helge Ritter
We establish two conditions which ensure the non-divergence of additive recurrent networks with unsaturating piecewise linear transfer functions, also called linear threshold or semilinear transfer...
Spatial feature binding and learning in competitive neural layer architectures / (2000)
Thesis (doctoral)--Universität, Bielefeld, 2000.
A Competitive Layer Model for Feature Binding and Sensory Segmentation (2000)
Heiko Wersing, Jochen J. Steil, Helge Ritter
We present a recurrent neural network for feature binding and sensory segmentation, the competitive layer model (CLM). The CLM uses topographically structured competitive and cooperative interactions...
A Neural Network Architecture for Automatic Segmentation of Fluorescence Micrographs (2000)
Tim W. Nattkemper, Heiko Wersing, Walter Schubert, Helge Ritter
. A system for the automatic segmentation of fluorescence micrographs is presented. In a first step positions of fluorescent cells are detected by a fast learning neural network, which acquires the...
Fluorescence Micrograph Segmentation by Gestalt-Based Feature Binding (2000)
Tim W. Nattkemper, Heiko Wersing, Walter Schubert, Helge Ritter
We present the application of a recurrent neural network feature binding model to the segmentation of uorescence micrographs, images showing uorescent cells in tonsil tissue. Image primitives,...
Feature binding and relaxation labeling with the competitive layer model (1999)
Abstract. We discuss the relation of the Competitive Layer Model (CLM) to Relaxation Labeling (RL) with regard to feature binding and labeling problems. The CLM uses cooperative and competitive...
Feature binding and relaxation labeling with the competitive layer model (1999)
Abstract. We discuss the relation of the Competitive Layer Model (CLM) to Relaxation Labeling (RL) with regard to feature binding and labeling problems. The CLM uses cooperative and competitive...
A layered recurrent neural network for feature grouping (1997)
Heiko Wersing, Jochen J. Steil, Helge Ritter
Abstract. We describe a recurrent network, the Competitive Layer Model (CLM) for feature grouping. The model uses a combination of cooperative and competitive interactions to partition a set of input...