Bridging Text Spotting and SLAM with Junction Features

Download: PDF.

“Bridging Text Spotting and SLAM with Junction Features” by H.-C. Wang, C. Finn, L. Paull, M. Kaess, R. Rosenholtz, S. Teller, and J.J. Leonard. In IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems, IROS, (Hamburg, Germany), Sep. 2015, pp. 3701-3708.

Abstract

Navigating in a previously unknown environment and recognizing naturally occurring text in a scene are two important autonomous capabilities that are typically treated as distinct. However, these two tasks are potentially complementary, (i) scene and pose priors can benefit text spotting, and (ii) the ability to identify and associate text features can benefit navigation accuracy through loop closures. Previous approaches to autonomous text spotting typically require significant training data and are too slow for real-time implementation. In this work, we propose a novel high-level feature descriptor, the "junction", which is particularly well-suited to text representation and is also fast to compute. We show that we are able to improve SLAM through text spotting on datasets collected with a Google Tango, illustrating how location priors enable improved loop closure with text features.

Download: PDF.

BibTeX entry:

@inproceedings{Wang15iros,
   author = {H.-C. Wang and C. Finn and L. Paull and M. Kaess and R.
	Rosenholtz and S. Teller and J.J. Leonard},
   title = {Bridging Text Spotting and {SLAM} with Junction Features},
   booktitle = {IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems, IROS},
   pages = {3701-3708},
   address = {Hamburg, Germany},
   month = sep,
   year = {2015}
}
Last updated: August 17, 2017