Condition-invariant, Top-down visual place recognition

Michael Milford, Walter Scheirer, Eleonora Vig, Arren Glover, Oliver Baumann, Jason Mattingley, David Cox

Research output: Contribution to journalArticleResearchpeer-review

21 Citations (Scopus)

Abstract

In This paper we present a novel, condition-invariant place recognition algorithm inspired by recent discoveries in human visual neuroscience. The algorithm combines intolerant but fast low resolution whole image matching with highly Tolerant, sub-image patch matching processes. The approach does not require prior Training and works on single images, alleviating The need for either a velocity signal or image sequence, differentiating it from current state of The art methods. We conduct an exhaustive set of experiments evaluating The relationship between place recognition performance and computational resources using part of The challenging Alderley sunny day - rainy night dataset, which has only been previously solved by integrating over 320 frame long image sequences. We achieve recall rates of up To 51% at 100% precision, matching places That have undergone drastic perceptual change while rejecting match hypotheses between highly aliased images of different places. Human Trials demonstrate The performance is approaching human capability. The results provide a new benchmark for single image, condition-invariant place recognition.

Original languageEnglish
Article number6907678
Pages (from-to)5571-5577
Number of pages7
JournalProceedings - IEEE International Conference on Robotics and Automation
DOIs
Publication statusPublished - 1 Jan 2014
Externally publishedYes
EventIEEE International Conference on Robotics and Automation - Hong Kong, China
Duration: 31 May 20147 Jun 2014
Conference number: 2014

Fingerprint

Image matching
Experiments

Cite this

Milford, Michael ; Scheirer, Walter ; Vig, Eleonora ; Glover, Arren ; Baumann, Oliver ; Mattingley, Jason ; Cox, David. / Condition-invariant, Top-down visual place recognition. In: Proceedings - IEEE International Conference on Robotics and Automation. 2014 ; pp. 5571-5577.
@article{82ca12b3ec8342de9f9d0a9bce25a292,
title = "Condition-invariant, Top-down visual place recognition",
abstract = "In This paper we present a novel, condition-invariant place recognition algorithm inspired by recent discoveries in human visual neuroscience. The algorithm combines intolerant but fast low resolution whole image matching with highly Tolerant, sub-image patch matching processes. The approach does not require prior Training and works on single images, alleviating The need for either a velocity signal or image sequence, differentiating it from current state of The art methods. We conduct an exhaustive set of experiments evaluating The relationship between place recognition performance and computational resources using part of The challenging Alderley sunny day - rainy night dataset, which has only been previously solved by integrating over 320 frame long image sequences. We achieve recall rates of up To 51{\%} at 100{\%} precision, matching places That have undergone drastic perceptual change while rejecting match hypotheses between highly aliased images of different places. Human Trials demonstrate The performance is approaching human capability. The results provide a new benchmark for single image, condition-invariant place recognition.",
author = "Michael Milford and Walter Scheirer and Eleonora Vig and Arren Glover and Oliver Baumann and Jason Mattingley and David Cox",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/ICRA.2014.6907678",
language = "English",
pages = "5571--5577",
journal = "Proceedings - IEEE International Conference on Robotics and Automation",
issn = "1050-4729",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",

}

Condition-invariant, Top-down visual place recognition. / Milford, Michael; Scheirer, Walter; Vig, Eleonora; Glover, Arren; Baumann, Oliver; Mattingley, Jason; Cox, David.

In: Proceedings - IEEE International Conference on Robotics and Automation, 01.01.2014, p. 5571-5577.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Condition-invariant, Top-down visual place recognition

AU - Milford, Michael

AU - Scheirer, Walter

AU - Vig, Eleonora

AU - Glover, Arren

AU - Baumann, Oliver

AU - Mattingley, Jason

AU - Cox, David

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In This paper we present a novel, condition-invariant place recognition algorithm inspired by recent discoveries in human visual neuroscience. The algorithm combines intolerant but fast low resolution whole image matching with highly Tolerant, sub-image patch matching processes. The approach does not require prior Training and works on single images, alleviating The need for either a velocity signal or image sequence, differentiating it from current state of The art methods. We conduct an exhaustive set of experiments evaluating The relationship between place recognition performance and computational resources using part of The challenging Alderley sunny day - rainy night dataset, which has only been previously solved by integrating over 320 frame long image sequences. We achieve recall rates of up To 51% at 100% precision, matching places That have undergone drastic perceptual change while rejecting match hypotheses between highly aliased images of different places. Human Trials demonstrate The performance is approaching human capability. The results provide a new benchmark for single image, condition-invariant place recognition.

AB - In This paper we present a novel, condition-invariant place recognition algorithm inspired by recent discoveries in human visual neuroscience. The algorithm combines intolerant but fast low resolution whole image matching with highly Tolerant, sub-image patch matching processes. The approach does not require prior Training and works on single images, alleviating The need for either a velocity signal or image sequence, differentiating it from current state of The art methods. We conduct an exhaustive set of experiments evaluating The relationship between place recognition performance and computational resources using part of The challenging Alderley sunny day - rainy night dataset, which has only been previously solved by integrating over 320 frame long image sequences. We achieve recall rates of up To 51% at 100% precision, matching places That have undergone drastic perceptual change while rejecting match hypotheses between highly aliased images of different places. Human Trials demonstrate The performance is approaching human capability. The results provide a new benchmark for single image, condition-invariant place recognition.

UR - http://www.scopus.com/inward/record.url?scp=84929224384&partnerID=8YFLogxK

U2 - 10.1109/ICRA.2014.6907678

DO - 10.1109/ICRA.2014.6907678

M3 - Article

SP - 5571

EP - 5577

JO - Proceedings - IEEE International Conference on Robotics and Automation

JF - Proceedings - IEEE International Conference on Robotics and Automation

SN - 1050-4729

M1 - 6907678

ER -