Learning spatial relations between objects from 3D scenes

Severin Fichtl, John Alexander, Frank Guerin, Wail Mustafa, Dirk Kraft, Norbert Kruger

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this work, we learn a limited number of abstractions which can then be used to form preconditions for motor actions. These abstractions take the form of spatial relations amongst objects. We consider three “classes” of spatial relation: The objects either are separated from, on-top of, or inside each other. We have tackled this same problem in previous work (Fichtl et al., 2013). Here we report on recent improved results using a novel application of histograms to visually recognise a spatial relation between objects in the environment. Using this histogram based approach we are able to report a very high rate of success when the system is asked to recognise a spatial relation.
Original languageEnglish
Title of host publication2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages1-2
Number of pages2
ISBN (Electronic)9781479910366
DOIs
Publication statusPublished - 2013
Event2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Osaka, Japan
Duration: 18 Aug 201322 Aug 2013

Conference

Conference2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013
CountryJapan
CityOsaka
Period18/08/1322/08/13

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Software

Cite this

Fichtl, S., Alexander, J., Guerin, F., Mustafa, W., Kraft, D., & Kruger, N. (2013). Learning spatial relations between objects from 3D scenes. In 2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings (pp. 1-2). [6652552] Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/DevLrn.2013.6652552

Learning spatial relations between objects from 3D scenes. / Fichtl, Severin; Alexander, John; Guerin, Frank; Mustafa, Wail; Kraft, Dirk; Kruger, Norbert.

2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings. Institute of Electrical and Electronics Engineers (IEEE), 2013. p. 1-2 6652552.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Fichtl, S, Alexander, J, Guerin, F, Mustafa, W, Kraft, D & Kruger, N 2013, Learning spatial relations between objects from 3D scenes. in 2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings., 6652552, Institute of Electrical and Electronics Engineers (IEEE), pp. 1-2, 2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013, Osaka, Japan, 18/08/13. https://doi.org/10.1109/DevLrn.2013.6652552
Fichtl S, Alexander J, Guerin F, Mustafa W, Kraft D, Kruger N. Learning spatial relations between objects from 3D scenes. In 2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings. Institute of Electrical and Electronics Engineers (IEEE). 2013. p. 1-2. 6652552 https://doi.org/10.1109/DevLrn.2013.6652552
Fichtl, Severin ; Alexander, John ; Guerin, Frank ; Mustafa, Wail ; Kraft, Dirk ; Kruger, Norbert. / Learning spatial relations between objects from 3D scenes. 2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings. Institute of Electrical and Electronics Engineers (IEEE), 2013. pp. 1-2
@inproceedings{92b33d5760aa45e1a08349eceb66a244,
title = "Learning spatial relations between objects from 3D scenes",
abstract = "In this work, we learn a limited number of abstractions which can then be used to form preconditions for motor actions. These abstractions take the form of spatial relations amongst objects. We consider three “classes” of spatial relation: The objects either are separated from, on-top of, or inside each other. We have tackled this same problem in previous work (Fichtl et al., 2013). Here we report on recent improved results using a novel application of histograms to visually recognise a spatial relation between objects in the environment. Using this histogram based approach we are able to report a very high rate of success when the system is asked to recognise a spatial relation.",
author = "Severin Fichtl and John Alexander and Frank Guerin and Wail Mustafa and Dirk Kraft and Norbert Kruger",
note = "This work was supported by the EU Cognitive Systems project XPERIENCE (FP7- ICT-270273) and Leverhulme Grant F/00 152/AL",
year = "2013",
doi = "10.1109/DevLrn.2013.6652552",
language = "English",
pages = "1--2",
booktitle = "2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings",
publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

}

TY - GEN

T1 - Learning spatial relations between objects from 3D scenes

AU - Fichtl, Severin

AU - Alexander, John

AU - Guerin, Frank

AU - Mustafa, Wail

AU - Kraft, Dirk

AU - Kruger, Norbert

N1 - This work was supported by the EU Cognitive Systems project XPERIENCE (FP7- ICT-270273) and Leverhulme Grant F/00 152/AL

PY - 2013

Y1 - 2013

N2 - In this work, we learn a limited number of abstractions which can then be used to form preconditions for motor actions. These abstractions take the form of spatial relations amongst objects. We consider three “classes” of spatial relation: The objects either are separated from, on-top of, or inside each other. We have tackled this same problem in previous work (Fichtl et al., 2013). Here we report on recent improved results using a novel application of histograms to visually recognise a spatial relation between objects in the environment. Using this histogram based approach we are able to report a very high rate of success when the system is asked to recognise a spatial relation.

AB - In this work, we learn a limited number of abstractions which can then be used to form preconditions for motor actions. These abstractions take the form of spatial relations amongst objects. We consider three “classes” of spatial relation: The objects either are separated from, on-top of, or inside each other. We have tackled this same problem in previous work (Fichtl et al., 2013). Here we report on recent improved results using a novel application of histograms to visually recognise a spatial relation between objects in the environment. Using this histogram based approach we are able to report a very high rate of success when the system is asked to recognise a spatial relation.

UR - http://www.scopus.com/inward/record.url?scp=84891085427&partnerID=8YFLogxK

U2 - 10.1109/DevLrn.2013.6652552

DO - 10.1109/DevLrn.2013.6652552

M3 - Conference contribution

AN - SCOPUS:84891085427

SP - 1

EP - 2

BT - 2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings

PB - Institute of Electrical and Electronics Engineers (IEEE)

ER -