Using deep autoencoders to investigate image matching in visual navigation

Walker, Christopher, Graham, Paul and Philippides, Andrew (2017) Using deep autoencoders to investigate image matching in visual navigation. Conference on Biomimetic and Biohybrid Systems. Published in: Biomimetic and Biohybrid Systems. Living Machines 2017. 10384 465-474. Springer, Cham ISSN 0302-9743 ISBN 9783319635361

[img] PDF - Accepted Version
Download (401kB)


This paper discusses the use of deep autoencoder networks to find a compressed representation of an image, which can be used for visual naviga-tion. Images reconstructed from the compressed representation are tested to see if they retain enough information to be used as a visual compass (in which an image is matched with another to recall a bearing/movement direction) as this ability is at the heart of a visual route navigation algorithm. We show that both reconstructed images and compressed representations from different layers of the autoencoder can be used in this way, suggesting that a compact image code is sufficient for visual navigation and that deep networks hold promise for find-ing optimal visual encodings for this task.

Item Type: Conference Proceedings
Keywords: Visual navigation, Insect-inspired robotics, Deep neural network, Autoencoder
Schools and Departments: School of Engineering and Informatics > Informatics
Research Centres and Groups: Centre for Computational Neuroscience and Robotics
Depositing User: Andy Philippides
Date Deposited: 21 Nov 2017 15:56
Last Modified: 03 Sep 2021 13:28

View download statistics for this item

📧 Request an update