Sökning: WFRF:(Adrian Gabriel) >
Learning Dense Visu...
Learning Dense Visual Descriptors using Image Augmentations for Robot Manipulation Tasks
-
- Graf, Christian (författare)
- Bosch Center for Artifical Intelligence, Bosch Center for Artifical Intelligence
-
- Adrian, David B. (författare)
- Bosch Center for Artifical Intelligence, Bosch Center for Artifical Intelligence; Ulm University, Germany
-
- Weil, Joshua (författare)
- KTH,Bosch Center for Artifical Intelligence, Bosch Center for Artifical Intelligence
-
visa fler...
-
- Gabriel, Miroslav (författare)
- Bosch Center for Artifical Intelligence, Bosch Center for Artifical Intelligence
-
- Schillinger, Philipp (författare)
- Bosch Center for Artifical Intelligence, Bosch Center for Artifical Intelligence
-
- Spies, Markus (författare)
- Bosch Center for Artifical Intelligence, Bosch Center for Artifical Intelligence
-
- Neumann, Heiko (författare)
- Ulm University, Germany
-
- Kupcsik, Andras (författare)
- Bosch Center for Artifical Intelligence, Bosch Center for Artifical Intelligence
-
visa färre...
-
(creator_code:org_t)
- ML Research Press, 2023
- 2023
- Engelska.
-
Ingår i: Proceedings of the 6th Conference on Robot Learning, CoRL 2022. - : ML Research Press. ; , s. 871-880
- Relaterad länk:
-
https://urn.kb.se/re...
Abstract
Ämnesord
Stäng
- We propose a self-supervised training approach for learning view-invariant dense visual descriptors using image augmentations. Unlike existing works, which often require complex datasets, such as registered RGBD sequences, we train on an unordered set of RGB images. This allows for learning from a single camera view, e.g., in an existing robotic cell with a fix-mounted camera. We create synthetic views and dense pixel correspondences using data augmentations. We find our approach to be competitive compared to existing methods, despite the simpler data recording and setup requirements. We show that training on synthetic correspondences provides descriptor consistency across a broad range of camera views. We compare against training with geometric correspondence from multiple views and provide ablation studies. We also show a robotic bin-picking experiment using descriptors learned from a fix-mounted camera for defining grasp preferences.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik -- Robotteknik och automation (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering -- Robotics (hsv//eng)
Nyckelord
- bin-picking
- computer vision
- representation learning
- self-supervised learning
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)