Learning Dense Visual Descriptors using Image Augmentations for Robot Manipulation Tasks
Learning Dense Visual Descriptors using Image Augmentations for Robot Manipulation Tasks
About this item
Full title
Author / Creator
Publisher
Ithaca: Cornell University Library, arXiv.org
Journal title
Language
English
Formats
Publication information
Publisher
Ithaca: Cornell University Library, arXiv.org
Subjects
More information
Scope and Contents
Contents
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 ce...
Alternative Titles
Full title
Learning Dense Visual Descriptors using Image Augmentations for Robot Manipulation Tasks
Authors, Artists and Contributors
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_2713856813
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2713856813
Other Identifiers
E-ISSN
2331-8422
