A self-organising network that grows when required

Marsland, Stephen; Shapiro, Johanna; Nehmzow, Ulrich

doi:10.1016/s0893-6080(02)00078-3

Cited by 352 publications

(305 citation statements)

References 22 publications

Supporting

Mentioning

292

Contrasting

Unclassified

Order By: Relevance

“…In this paper, we use a growing-when-required (GWR) network [15] for efficient clustering of keypoints. When performing 3D object recognition, many of the acquired keypoints look very similar.…”

Section: Active Exploration and Keypoint Clustering For Object Recognmentioning

confidence: 99%

“…We therefore use a growing-when-required (GWR) network [15] to efficiently cluster keypoints that are highly similar. A GWR-network is a clustering method, very similar to a growing-neural-gas (GNG) network [7].…”

Section: Keypoint Clusteringmentioning

confidence: 99%

“…The GWR-network as described in [15] uses edges between nodes. This is based upon the SOM, and results in a topological preservation of the network in the sense that connected nodes in the network correspond to neighboring points in the input space.…”

Section: Keypoint Clusteringmentioning

confidence: 99%

“…For the description of our implementation of the GWRnetwork, we follow the notation and description in [15]. Let K be the set of observed keypoints when learning the objects, A be the set of nodes in the network, n w be the weight vector of node n (of the same dimensionality as the SIFT keypoints), and n t be the activation counter.…”

Section: Keypoint Clusteringmentioning

confidence: 99%

See 3 more Smart Citations

Active exploration and keypoint clustering for object recognition

Kootstra

Ypma

Boer

2008

2008 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

Abstract-Object recognition is a challenging problem for artificial systems. This is especially true for objects that are placed in cluttered and uncontrolled environments. To challenge this problem, we discuss an active approach to object recognition. Instead of passively observing objects, we use a robot to actively explore the objects. This enables the system to learn objects from different viewpoints and to actively select viewpoints for optimal recognition. Active vision furthermore simplifies the segmentation of the object from its background. As the basis for object recognition we use the Scale Invariant Feature Transform (SIFT). SIFT has been a successful method for image representation. However, a known drawback of SIFT is that the computational complexity of the algorithm increases with the number of keypoints. We discuss a growing-whenrequired (GWR) network for efficient clustering of the keypoints. The results show successful learning of 3D objects in real-world environments. The active approach is successful in separating the object from its cluttered background, and the active selection of viewpoint further increases the performance. Moreover, the GWR-network strongly reduces the number of keypoints.

show abstract

Section: Active Exploration and Keypoint Clustering For Object Recognmentioning

confidence: 99%

Section: Keypoint Clusteringmentioning

confidence: 99%

Section: Keypoint Clusteringmentioning

confidence: 99%

Section: Keypoint Clusteringmentioning

confidence: 99%

See 2 more Smart Citations

Active exploration and keypoint clustering for object recognition

Kootstra

Ypma

Boer

2008

2008 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

show abstract

“…This also applied to the method proposed in (Marsland et al, 2002), where new nodes can be added at any time during the learning process, which are positioned based on the input and the current winning node, rather than adding them where the accumulated error is maximized, which is the case with GNG.…”

Section: Introductionmentioning

confidence: 99%

Growing Neural Gas approach for obtaining homogeneous maps by restricting the insertion of new nodes

2014

View full text Add to dashboard Cite

The Growing Neural Gas model is used widely in artificial neural networks.However, its application is limited in some contexts by the proliferation of nodes in dense areas of the input space. In this study, we introduce some modifications to address this problem by imposing three restrictions on the insertion of new nodes. Each restriction aims to maintain the homogeneous values of selected criteria. One criterion is related to the square error of classification and an alternative approach is proposed for avoiding additional computational costs.Three parameters are added that allow the regulation of the restriction criteria.The resulting algorithm allows models to be obtained that suit specific needs by specifying meaningful parameters.

show abstract

Evolving Intelligent Systems

Baruah

Angelov

Baruah

2020

Wiley Encyclopedia of Electrical and Electronics Engineering

View full text Add to dashboard Cite

This article provides an overview of Evolving Intelligent Systems. It discusses the architecture and design of evolving frameworks that are based on computational intelligence, namely, evolving fuzzy systems, evolving neural networks, and evolving neuro‐fuzzy systems. The article further describes evolving approaches in the context of unsupervised and supervised learning. It highlights several successful applications of evolving systems. The reader would also be introduced to the novel concept of collaborative evolving intelligent systems through this article. Finally, the article is concluded with future research directions.

show abstract

A self-organising network that grows when required

Cited by 352 publications

References 22 publications

Active exploration and keypoint clustering for object recognition

Active exploration and keypoint clustering for object recognition

Growing Neural Gas approach for obtaining homogeneous maps by restricting the insertion of new nodes

Evolving Intelligent Systems

Contact Info

Product

Resources

About