Improving robot vision models for object detection through interaction

Jurgen Leitner; Alexander Forster; Jurgen Schmidhuber

doi:10.1109/IJCNN.2014.6889556

Improving robot vision models for object detection through interaction

Jurgen Leitner, Alexander Forster, Jurgen Schmidhuber

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

8 Citations (Scopus)

Abstract

We propose a method for learning specific object representations that can be applied (and reused) in visual detection and identification tasks. A machine learning technique called Cartesian Genetic Programming (CGP) is used to create these models based on a series of images. Our research investigates how manipulation actions might allow for the development of better visual models and therefore better robot vision. This paper describes how visual object representations can be learned and improved by performing object manipulation actions, such as, poke, push and pick-up with a humanoid robot. The improvement can be measured and allows for the robot to select and perform the right action, i.e. the action with the best possible improvement of the detector.

Original language	English
Title of host publication	Proceedings of the International Joint Conference on Neural Networks
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	3355-3362
Number of pages	8
ISBN (Electronic)	9781479914845
DOIs	https://doi.org/10.1109/IJCNN.2014.6889556
Publication status	Published - 3 Sept 2014
Externally published	Yes
Event	IEEE International Joint Conference on Neural Networks 2014 - Beijing, China Duration: 6 Jul 2014 → 11 Jul 2014 https://ieeexplore.ieee.org/xpl/conhome/6880678/proceeding (Proceedings)

Conference

Conference	IEEE International Joint Conference on Neural Networks 2014
Abbreviated title	IJCNN 2014
Country/Territory	China
City	Beijing
Period	6/07/14 → 11/07/14
Internet address	https://ieeexplore.ieee.org/xpl/conhome/6880678/proceeding (Proceedings)

Access to Document

10.1109/IJCNN.2014.6889556

Cite this

@inproceedings{954cedab4564473cb252be39f3718d5d,

title = "Improving robot vision models for object detection through interaction",

abstract = "We propose a method for learning specific object representations that can be applied (and reused) in visual detection and identification tasks. A machine learning technique called Cartesian Genetic Programming (CGP) is used to create these models based on a series of images. Our research investigates how manipulation actions might allow for the development of better visual models and therefore better robot vision. This paper describes how visual object representations can be learned and improved by performing object manipulation actions, such as, poke, push and pick-up with a humanoid robot. The improvement can be measured and allows for the robot to select and perform the right action, i.e. the action with the best possible improvement of the detector.",

author = "Jurgen Leitner and Alexander Forster and Jurgen Schmidhuber",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; IEEE International Joint Conference on Neural Networks 2014, IJCNN 2014 ; Conference date: 06-07-2014 Through 11-07-2014",

year = "2014",

month = sep,

day = "3",

doi = "10.1109/IJCNN.2014.6889556",

language = "English",

pages = "3355--3362",

booktitle = "Proceedings of the International Joint Conference on Neural Networks",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

url = "https://ieeexplore.ieee.org/xpl/conhome/6880678/proceeding",

}

Leitner, J, Forster, A & Schmidhuber, J 2014, Improving robot vision models for object detection through interaction. in Proceedings of the International Joint Conference on Neural Networks. IEEE, Institute of Electrical and Electronics Engineers, pp. 3355-3362, IEEE International Joint Conference on Neural Networks 2014, Beijing, China, 6/07/14. https://doi.org/10.1109/IJCNN.2014.6889556

Improving robot vision models for object detection through interaction. / Leitner, Jurgen; Forster, Alexander; Schmidhuber, Jurgen.
Proceedings of the International Joint Conference on Neural Networks. IEEE, Institute of Electrical and Electronics Engineers, 2014. p. 3355-3362.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Improving robot vision models for object detection through interaction

AU - Leitner, Jurgen

AU - Forster, Alexander

AU - Schmidhuber, Jurgen

PY - 2014/9/3

Y1 - 2014/9/3

N2 - We propose a method for learning specific object representations that can be applied (and reused) in visual detection and identification tasks. A machine learning technique called Cartesian Genetic Programming (CGP) is used to create these models based on a series of images. Our research investigates how manipulation actions might allow for the development of better visual models and therefore better robot vision. This paper describes how visual object representations can be learned and improved by performing object manipulation actions, such as, poke, push and pick-up with a humanoid robot. The improvement can be measured and allows for the robot to select and perform the right action, i.e. the action with the best possible improvement of the detector.

AB - We propose a method for learning specific object representations that can be applied (and reused) in visual detection and identification tasks. A machine learning technique called Cartesian Genetic Programming (CGP) is used to create these models based on a series of images. Our research investigates how manipulation actions might allow for the development of better visual models and therefore better robot vision. This paper describes how visual object representations can be learned and improved by performing object manipulation actions, such as, poke, push and pick-up with a humanoid robot. The improvement can be measured and allows for the robot to select and perform the right action, i.e. the action with the best possible improvement of the detector.

UR - http://www.scopus.com/inward/record.url?scp=84908494281&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2014.6889556

DO - 10.1109/IJCNN.2014.6889556

M3 - Conference Paper

AN - SCOPUS:84908494281

SP - 3355

EP - 3362

BT - Proceedings of the International Joint Conference on Neural Networks

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - IEEE International Joint Conference on Neural Networks 2014

Y2 - 6 July 2014 through 11 July 2014

ER -

Improving robot vision models for object detection through interaction

Abstract

Conference

Access to Document

Other files and links

Cite this