Birth of the Object: Detection of Objectness and Extraction of Object Shape through Object Action Complexes

Dirk Kraft, Nicolas Pugeault, Emre Baseski, Mila Popovic, Danica Kragic, Sinan Kalkan, Florentin Wörgötter, Norbert Krüger

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

We describe a process in which the segmentation of objects as well as the extraction of the object shape becomes realized through active exploration of a robot vision system. In the exploration process, two behavioral modules that link robot actions to the visual and haptic perception of objects interact. First, by making use of an object independent grasping mechanism, physical control over potential objects can be gained. Having evaluated the initial grasping mechanism as being successful, a second behavior extracts the object shape by making use of prediction based on the motion induced by the robot. This also leads to the concept of an "object" as a set of features that change predictably over different frames.

The system is equipped with a certain degree of generic prior knowledge about the world in terms of a sophisticated visual feature extraction process in an early cognitive vision system, knowledge about its own embodiment as well as knowledge about geometric relationships such as rigid body motion. This prior knowledge allows the extraction of representations that are semantically richer compared to many other approaches.
OriginalsprogEngelsk
TidsskriftInternational Journal of Humanoid Robotics
Vol/bind5
Udgave nummer2
Sider (fra-til)247-265
ISSN0219-8436
DOI
StatusUdgivet - 2008

Fingeraftryk

Robots
Computer vision
Feature extraction
Object detection

Citer dette

Kraft, Dirk ; Pugeault, Nicolas ; Baseski, Emre ; Popovic, Mila ; Kragic, Danica ; Kalkan, Sinan ; Wörgötter, Florentin ; Krüger, Norbert. / Birth of the Object: Detection of Objectness and Extraction of Object Shape through Object Action Complexes. I: International Journal of Humanoid Robotics. 2008 ; Bind 5, Nr. 2. s. 247-265.
@article{4e799c60e5e011dc9a76000ea68e967b,
title = "Birth of the Object: Detection of Objectness and Extraction of Object Shape through Object Action Complexes",
abstract = "We describe a process in which the segmentation of objects as well as the extraction of the object shape becomes realized through active exploration of a robot vision system. In the exploration process, two behavioral modules that link robot actions to the visual and haptic perception of objects interact. First, by making use of an object independent grasping mechanism, physical control over potential objects can be gained. Having evaluated the initial grasping mechanism as being successful, a second behavior extracts the object shape by making use of prediction based on the motion induced by the robot. This also leads to the concept of an {"}object{"} as a set of features that change predictably over different frames.The system is equipped with a certain degree of generic prior knowledge about the world in terms of a sophisticated visual feature extraction process in an early cognitive vision system, knowledge about its own embodiment as well as knowledge about geometric relationships such as rigid body motion. This prior knowledge allows the extraction of representations that are semantically richer compared to many other approaches.",
keywords = "Early cognitive vision, grasping, exploration",
author = "Dirk Kraft and Nicolas Pugeault and Emre Baseski and Mila Popovic and Danica Kragic and Sinan Kalkan and Florentin W{\"o}rg{\"o}tter and Norbert Kr{\"u}ger",
year = "2008",
doi = "10.1142/S021984360800139X",
language = "English",
volume = "5",
pages = "247--265",
journal = "International Journal of Humanoid Robotics",
issn = "0219-8436",
publisher = "World Scientific",
number = "2",

}

Birth of the Object: Detection of Objectness and Extraction of Object Shape through Object Action Complexes. / Kraft, Dirk; Pugeault, Nicolas; Baseski, Emre; Popovic, Mila; Kragic, Danica; Kalkan, Sinan; Wörgötter, Florentin; Krüger, Norbert.

I: International Journal of Humanoid Robotics, Bind 5, Nr. 2, 2008, s. 247-265.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Birth of the Object: Detection of Objectness and Extraction of Object Shape through Object Action Complexes

AU - Kraft, Dirk

AU - Pugeault, Nicolas

AU - Baseski, Emre

AU - Popovic, Mila

AU - Kragic, Danica

AU - Kalkan, Sinan

AU - Wörgötter, Florentin

AU - Krüger, Norbert

PY - 2008

Y1 - 2008

N2 - We describe a process in which the segmentation of objects as well as the extraction of the object shape becomes realized through active exploration of a robot vision system. In the exploration process, two behavioral modules that link robot actions to the visual and haptic perception of objects interact. First, by making use of an object independent grasping mechanism, physical control over potential objects can be gained. Having evaluated the initial grasping mechanism as being successful, a second behavior extracts the object shape by making use of prediction based on the motion induced by the robot. This also leads to the concept of an "object" as a set of features that change predictably over different frames.The system is equipped with a certain degree of generic prior knowledge about the world in terms of a sophisticated visual feature extraction process in an early cognitive vision system, knowledge about its own embodiment as well as knowledge about geometric relationships such as rigid body motion. This prior knowledge allows the extraction of representations that are semantically richer compared to many other approaches.

AB - We describe a process in which the segmentation of objects as well as the extraction of the object shape becomes realized through active exploration of a robot vision system. In the exploration process, two behavioral modules that link robot actions to the visual and haptic perception of objects interact. First, by making use of an object independent grasping mechanism, physical control over potential objects can be gained. Having evaluated the initial grasping mechanism as being successful, a second behavior extracts the object shape by making use of prediction based on the motion induced by the robot. This also leads to the concept of an "object" as a set of features that change predictably over different frames.The system is equipped with a certain degree of generic prior knowledge about the world in terms of a sophisticated visual feature extraction process in an early cognitive vision system, knowledge about its own embodiment as well as knowledge about geometric relationships such as rigid body motion. This prior knowledge allows the extraction of representations that are semantically richer compared to many other approaches.

KW - Early cognitive vision

KW - grasping

KW - exploration

U2 - 10.1142/S021984360800139X

DO - 10.1142/S021984360800139X

M3 - Journal article

VL - 5

SP - 247

EP - 265

JO - International Journal of Humanoid Robotics

JF - International Journal of Humanoid Robotics

SN - 0219-8436

IS - 2

ER -