Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/117928
Title: Learning depth-aware deep representations for robotic perception
Authors: Porzi, Lorenzo
Rota Buló, Samuel
Penate-Sanchez, Adrian 
Ricci, Elisa
Moreno-Noguer, Frances
Keywords: RGB-D perception
Visual learning
Issue Date: 2017
Journal: IEEE Robotics and Automation Letters 
Abstract: Exploiting RGB-D data by means of convolutional neural networks (CNNs) is at the core of a number of robotics applications, including object detection, scene semantic segmentation, and grasping. Most existing approaches, however, exploit RGB-D data by simply considering depth as an additional input channel for the network. In this paper we show that the performance of deep architectures can be boosted by introducing DaConv, a novel, general-purpose CNN block which exploits depth to learn scale-aware feature representations. We demonstrate the benefits of DaConv on a variety of robotics oriented tasks, involving affordance detection, object coordinate regression, and contour detection in RGB-D images. In each of these experiments we show the potential of the proposed block and how it can be readily integrated into existing CNN architectures.
URI: http://hdl.handle.net/10553/117928
ISSN: 2377-3766
DOI: 10.1109/LRA.2016.2637444
Source: IEEE Robotics and Automation Letters, v. 2 (2), p. 468 - 475 (1997)
Appears in Collections:Artículos
Unknown (1,38 MB)
Show full item record

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.