Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas

Kortkamp M, Wachsmuth S (2010)
In: International Conference on Artificial Neural Networks. Thessaloniki, Greece: Springer: 188-197.

Konferenzbeitrag | Veröffentlicht | Englisch
 
Download
Es wurden keine Dateien hochgeladen. Nur Publikationsnachweis!
Abstract / Bemerkung
The Growing Neural Gas (GNG) algorithm is able to perform continuous vector quantization for an unknown distribution while preserving the topological structure of the input space. This makes the GNG attractive for online learning of visual codebooks. However, mapping an input vector to a reference vector is quite expensive and requires an iteration through the entire codebook. We propose a hierarchical extension of the Growing Neural Gas algorithm for online one-shot learning of visual vocabularies. The method intrinsically supports mapping input vectors to codewords in sub-linear time. Further, our extension avoids overfitting and locally keeps track of the topology of the input space. The algorithm is evaluated on both, low dimensional simulated data and high dimensional real world data.
Stichworte
GNG; patterns; unsupervised; visual; online; one-shot; learning; lbTreeGNG; hierarchical; codewords; constructive
Erscheinungsjahr
2010
Titel des Konferenzbandes
International Conference on Artificial Neural Networks
Seite(n)
188-197
Page URI
https://pub.uni-bielefeld.de/record/2034693

Zitieren

Kortkamp M, Wachsmuth S. Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas. In: International Conference on Artificial Neural Networks. Thessaloniki, Greece: Springer; 2010: 188-197.
Kortkamp, M., & Wachsmuth, S. (2010). Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas. International Conference on Artificial Neural Networks, 188-197. Thessaloniki, Greece: Springer.
Kortkamp, Marco, and Wachsmuth, Sven. 2010. “Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas”. In International Conference on Artificial Neural Networks, 188-197. Thessaloniki, Greece: Springer.
Kortkamp, M., and Wachsmuth, S. (2010). “Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas” in International Conference on Artificial Neural Networks (Thessaloniki, Greece: Springer), 188-197.
Kortkamp, M., & Wachsmuth, S., 2010. Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas. In International Conference on Artificial Neural Networks. Thessaloniki, Greece: Springer, pp. 188-197.
M. Kortkamp and S. Wachsmuth, “Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas”, International Conference on Artificial Neural Networks, Thessaloniki, Greece: Springer, 2010, pp.188-197.
Kortkamp, M., Wachsmuth, S.: Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas. International Conference on Artificial Neural Networks. p. 188-197. Springer, Thessaloniki, Greece (2010).
Kortkamp, Marco, and Wachsmuth, Sven. “Continuous Visual Codebooks with a Limited Branching Tree Growing Neural Gas”. International Conference on Artificial Neural Networks. Thessaloniki, Greece: Springer, 2010. 188-197.
Export

Markieren/ Markierung löschen
Markierte Publikationen

Open Data PUB

Suchen in

Google Scholar