A Deep Learning Framework for Tactile Recognition of Known as well as Novel Objects

Zineb Abderrahmane 1 Gowrishankar Ganesh 2, 1 André Crosnier 1 Andrea Cherubini 1
1 IDH - Interactive Digital Humans
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : This paper addresses the recognition of daily-life objects by a robot equipped with tactile sensors. The main contribution is a deep learning framework that can recognize objects already touched as well as objects never touched before. To this end, we train a Deconvolutional Neural Network that generates synthetic tactile data for novel classes. Then, we use both these synthetic data and the real data collected by touching objects, to train a Convolutional Neural Network to recognize both known (trained) objects and novel objects. Furthermore, we propose a method for integrating newly encountered data into novel classes. Finally, we evaluate the framework using the largest available dataset of tactile objects descriptions.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02012628
Contributor : Andrea Cherubini <>
Submitted on : Friday, February 8, 2019 - 7:30:03 PM
Last modification on : Thursday, May 16, 2019 - 12:48:01 PM
Long-term archiving on : Thursday, May 9, 2019 - 4:54:42 PM

File

Deep_TII_zineb_final.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Zineb Abderrahmane, Gowrishankar Ganesh, André Crosnier, Andrea Cherubini. A Deep Learning Framework for Tactile Recognition of Known as well as Novel Objects. IEEE Transactions on Industrial Informatics, Institute of Electrical and Electronics Engineers, In press, ⟨10.1109/TII.2019.2898264⟩. ⟨hal-02012628⟩

Share

Metrics

Record views

87

Files downloads

130