Integrating feature maps and competitive layer architectures for motion segmentation

Authors:
Jan Steffen;Michael Pardowitz;Jochen J. Steil;Helge Ritter
Affiliations:
Neuroinformatics Group, Faculty of Technology, Bielefeld University, Germany;Neuroinformatics Group, Faculty of Technology, Bielefeld University, Germany and Research Institute for Cognition and Robotics (CoR-Lab), Bielefeld University, Germany;Research Institute for Cognition and Robotics (CoR-Lab), Bielefeld University, Germany;Neuroinformatics Group, Faculty of Technology, Bielefeld University, Germany and Research Institute for Cognition and Robotics (CoR-Lab), Bielefeld University, Germany
Venue:
Neurocomputing
Year:
2011

Citing 16
Cited 0

Self-Organizing Maps

Self-Organizing Maps
Selectively grouping neurons in recurrent networks of lateral inhibition

Neural Computation
Multistability analysis for recurrent neural networks with unsaturating piecewise linear transfer functions

Neural Computation
Principal Surfaces from Unsupervised Kernel Regression

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Competitive-Layer Model for Feature Binding and Sensory Segmentation

Neural Computation
Perspectives of Neural-Symbolic Integration

Perspectives of Neural-Symbolic Integration
Towards Semi-supervised Manifold Learning: UKR with Structural Hints

WSOM '09 Proceedings of the 7th International Workshop on Advances in Self-Organizing Maps
Solving the CLM Problem by Discrete-Time Linear Threshold Recurrent Neural Networks

ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part I
Using structured UKR manifolds for motion classification and segmentation

IROS'09 Proceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems
Learning compatibility functions for feature binding and perceptual grouping

ICANN/ICONIP'03 Proceedings of the 2003 joint international conference on Artificial neural networks and neural information processing
Self-emerging action gestalts for task segmentation

KI'09 Proceedings of the 32nd annual German conference on Advances in artificial intelligence
Foundations of implementing the competitive layer model by Lotka-Volterra recurrent neural networks

IEEE Transactions on Neural Networks
Competitive layer model of discrete-time recurrent neural networks with lt neurons

Neural Computation
A leave-k-out cross-validation scheme for unsupervised kernel regression

ICANN'06 Proceedings of the 16th international conference on Artificial Neural Networks - Volume Part II
Dynamics analysis and analog associative memory of networks with LT neurons

IEEE Transactions on Neural Networks
Learning lateral interactions for feature binding and sensory segmentation from prototypic basis interactions

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.01

Visualization

Abstract

We present a generic approach to integrate feature maps with a competitive layer architecture to enable segmentation by a competitive neural dynamics specified in terms of the latent space mappings constructed by the feature maps. We demonstrate the underlying ideas for the case of motion segmentation, using a system that employs Unsupervised Kernel Regression (UKR) for the creation of the feature maps, and the Competitive Layer Model (CLM) for the competitive layer architecture. The UKR feature maps hold learned representations of a set of candidate motions and the CLM dynamics, working on features defined in the UKR domain, implements the segmentation of observed trajectory data according to the competing candidates. We also demonstrate how the introduction of an additional layer can provide the system with a parametrizable rejection mechanism for previously unknown observations. The evaluation on trajectories describing four different letters yields improved classification results compared to our previous, pure manifold approach.