From Geometries to Contact Graphs

Meier M, Haschke R, Ritter H (2020)
In: Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II. Farkas I, Masulli P, Wermter S (Eds); Lecture Notes in Computer Science, 12397. Cham: Springer: 546-555.

Konferenzbeitrag | Veröffentlicht | Englisch
 
Download
OA 1.89 MB
Herausgeber*in
Farkas, Igor; Masulli, Paolo; Wermter, Stefan
Abstract / Bemerkung
When a robot perceives its environment, it is not only important to know what kind of objects are present in it, but also how they relate to each other. For example in a cleanup task in a cluttered environment, a sensible strategy is to pick the objects with the least contacts to other objects first, to minimize the chance of unwanted movements not related to the current picking action. Estimating object contacts in cluttered scenes only based on passive observation is a complex problem. To tackle this problem, we present a deep neural network that learns physically stable object relations directly from geometric features. The learned relations are encoded as contact graphs between the objects. To facilitate training of the network, we generated a rich, publicly available dataset consisting of more than 25000 unique contact scenes, by utilizing a physics simulation. Different deep architectures have been evaluated and the final architecture, which shows good results in reconstructing contact graphs, is evaluated quantitatively and qualitatively.
Erscheinungsjahr
2020
Titel des Konferenzbandes
Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II
Band
12397
Seite(n)
546-555
Konferenz
International Conference on Artificial Neural Networks
Konferenzort
Bratislava, Slovakia
Konferenzdatum
2020-09-15 – 2020-09-18
ISBN
978-3-030-61615-1
Page URI
https://pub.uni-bielefeld.de/record/2945542

Zitieren

Meier M, Haschke R, Ritter H. From Geometries to Contact Graphs. In: Farkas I, Masulli P, Wermter S, eds. Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II. Lecture Notes in Computer Science. Vol 12397. Cham: Springer; 2020: 546-555.
Meier, M., Haschke, R., & Ritter, H. (2020). From Geometries to Contact Graphs. In I. Farkas, P. Masulli, & S. Wermter (Eds.), Lecture Notes in Computer Science: Vol. 12397. Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II (pp. 546-555). Cham: Springer. doi:10.1007/978-3-030-61616-8_44
Meier, M., Haschke, R., and Ritter, H. (2020). “From Geometries to Contact Graphs” in Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II, Farkas, I., Masulli, P., and Wermter, S. eds. Lecture Notes in Computer Science, vol. 12397, (Cham: Springer), 546-555.
Meier, M., Haschke, R., & Ritter, H., 2020. From Geometries to Contact Graphs. In I. Farkas, P. Masulli, & S. Wermter, eds. Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II. Lecture Notes in Computer Science. no.12397 Cham: Springer, pp. 546-555.
M. Meier, R. Haschke, and H. Ritter, “From Geometries to Contact Graphs”, Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II, I. Farkas, P. Masulli, and S. Wermter, eds., Lecture Notes in Computer Science, vol. 12397, Cham: Springer, 2020, pp.546-555.
Meier, M., Haschke, R., Ritter, H.: From Geometries to Contact Graphs. In: Farkas, I., Masulli, P., and Wermter, S. (eds.) Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II. Lecture Notes in Computer Science. 12397, p. 546-555. Springer, Cham (2020).
Meier, Martin, Haschke, Robert, and Ritter, Helge. “From Geometries to Contact Graphs”. Artificial Neural Networks and Machine Learning – ICANN 2020. Proceedings. Part II. Ed. Igor Farkas, Paolo Masulli, and Stefan Wermter. Cham: Springer, 2020.Vol. 12397. Lecture Notes in Computer Science. 546-555.
Alle Dateien verfügbar unter der/den folgenden Lizenz(en):
Creative Commons Namensnennung 4.0 International Public License (CC-BY 4.0):
Volltext(e)
Name
Access Level
OA Open Access
Zuletzt Hochgeladen
2021-02-01T10:42:07Z
MD5 Prüfsumme
44d500f9fb50629e432cc39ce881168d

Export

Markieren/ Markierung löschen
Markierte Publikationen

Open Data PUB

Suchen in

Google Scholar
ISBN Suche