Affordance estimation for vision-based object replacement on a humanoid robot

Wail Mustafa, Mirko Wächter, Sandor Szedmak, Alejandro Agostini, Dirk Kraft, Tamim Asfour, Justus Piater, Florentin Wörgötter, Norbert Krüger

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

Resumé

In this paper, we address the problem of finding replacements of missing objects, involved in the execution of manipulation tasks. Our approach is based on estimating functional affordances for the unknown objects in order to propose replacements. We use a vision-based affordance estimation system utilizing object-wise global features and a multi-label learning method. This method also associates confidence values to the estimated affordances. We evaluate our approach on kitchen-related manipulation affordances. The evaluation also includes testing different scenarios for training the system using large-scale datasets. The results indicate that the system is able to successfully predict the affordances of novel objects. We also implement our system on a humanoid robot and demonstrate the affordance estimation in a real scene.

OriginalsprogEngelsk
TitelProceedings of the 47th International Symposium on Robotics
ForlagVDE-Verlag
Publikationsdato2016
Sider164-172
ISBN (Trykt)978-3-8007-4231-8
StatusUdgivet - 2016
Begivenhed47th International Symposium on Robotics - Munich, Tyskland
Varighed: 21. jun. 201622. jun. 2016
Konferencens nummer: 47

Konference

Konference47th International Symposium on Robotics
Nummer47
LandTyskland
ByMunich
Periode21/06/201622/06/2016
SponsorABB Robotics, FANUC Deutschland GmbH, KUKA Roboter GmbH, SCHUNK GmbH and Co. KG

Fingeraftryk

Robots
Kitchens
Large scale systems
Labels
Testing

Citer dette

Mustafa, W., Wächter, M., Szedmak, S., Agostini, A., Kraft, D., Asfour, T., ... Krüger, N. (2016). Affordance estimation for vision-based object replacement on a humanoid robot. I Proceedings of the 47th International Symposium on Robotics (s. 164-172). VDE-Verlag.
Mustafa, Wail ; Wächter, Mirko ; Szedmak, Sandor ; Agostini, Alejandro ; Kraft, Dirk ; Asfour, Tamim ; Piater, Justus ; Wörgötter, Florentin ; Krüger, Norbert. / Affordance estimation for vision-based object replacement on a humanoid robot. Proceedings of the 47th International Symposium on Robotics. VDE-Verlag, 2016. s. 164-172
@inproceedings{c6fb5c152d31434788c9f48175da2182,
title = "Affordance estimation for vision-based object replacement on a humanoid robot",
abstract = "In this paper, we address the problem of finding replacements of missing objects, involved in the execution of manipulation tasks. Our approach is based on estimating functional affordances for the unknown objects in order to propose replacements. We use a vision-based affordance estimation system utilizing object-wise global features and a multi-label learning method. This method also associates confidence values to the estimated affordances. We evaluate our approach on kitchen-related manipulation affordances. The evaluation also includes testing different scenarios for training the system using large-scale datasets. The results indicate that the system is able to successfully predict the affordances of novel objects. We also implement our system on a humanoid robot and demonstrate the affordance estimation in a real scene.",
author = "Wail Mustafa and Mirko W{\"a}chter and Sandor Szedmak and Alejandro Agostini and Dirk Kraft and Tamim Asfour and Justus Piater and Florentin W{\"o}rg{\"o}tter and Norbert Kr{\"u}ger",
year = "2016",
language = "English",
isbn = "978-3-8007-4231-8",
pages = "164--172",
booktitle = "Proceedings of the 47th International Symposium on Robotics",
publisher = "VDE-Verlag",

}

Mustafa, W, Wächter, M, Szedmak, S, Agostini, A, Kraft, D, Asfour, T, Piater, J, Wörgötter, F & Krüger, N 2016, Affordance estimation for vision-based object replacement on a humanoid robot. i Proceedings of the 47th International Symposium on Robotics. VDE-Verlag, s. 164-172, 47th International Symposium on Robotics, Munich, Tyskland, 21/06/2016.

Affordance estimation for vision-based object replacement on a humanoid robot. / Mustafa, Wail; Wächter, Mirko; Szedmak, Sandor; Agostini, Alejandro; Kraft, Dirk; Asfour, Tamim; Piater, Justus; Wörgötter, Florentin; Krüger, Norbert.

Proceedings of the 47th International Symposium on Robotics. VDE-Verlag, 2016. s. 164-172.

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

TY - GEN

T1 - Affordance estimation for vision-based object replacement on a humanoid robot

AU - Mustafa, Wail

AU - Wächter, Mirko

AU - Szedmak, Sandor

AU - Agostini, Alejandro

AU - Kraft, Dirk

AU - Asfour, Tamim

AU - Piater, Justus

AU - Wörgötter, Florentin

AU - Krüger, Norbert

PY - 2016

Y1 - 2016

N2 - In this paper, we address the problem of finding replacements of missing objects, involved in the execution of manipulation tasks. Our approach is based on estimating functional affordances for the unknown objects in order to propose replacements. We use a vision-based affordance estimation system utilizing object-wise global features and a multi-label learning method. This method also associates confidence values to the estimated affordances. We evaluate our approach on kitchen-related manipulation affordances. The evaluation also includes testing different scenarios for training the system using large-scale datasets. The results indicate that the system is able to successfully predict the affordances of novel objects. We also implement our system on a humanoid robot and demonstrate the affordance estimation in a real scene.

AB - In this paper, we address the problem of finding replacements of missing objects, involved in the execution of manipulation tasks. Our approach is based on estimating functional affordances for the unknown objects in order to propose replacements. We use a vision-based affordance estimation system utilizing object-wise global features and a multi-label learning method. This method also associates confidence values to the estimated affordances. We evaluate our approach on kitchen-related manipulation affordances. The evaluation also includes testing different scenarios for training the system using large-scale datasets. The results indicate that the system is able to successfully predict the affordances of novel objects. We also implement our system on a humanoid robot and demonstrate the affordance estimation in a real scene.

M3 - Article in proceedings

AN - SCOPUS:84983776379

SN - 978-3-8007-4231-8

SP - 164

EP - 172

BT - Proceedings of the 47th International Symposium on Robotics

PB - VDE-Verlag

ER -

Mustafa W, Wächter M, Szedmak S, Agostini A, Kraft D, Asfour T et al. Affordance estimation for vision-based object replacement on a humanoid robot. I Proceedings of the 47th International Symposium on Robotics. VDE-Verlag. 2016. s. 164-172