Place recognition from disparate views

Rob Frampton; Andrew D Calway

Place recognition from disparate views

Rob Frampton
, Andrew D Calway

Visual Information Laboratory

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

73 Downloads (Pure)

Abstract

We use detected objects as basic features in a semantic place recognition system with the aim of allowing recognition when views are disparate. This is achieved by constructing a 2D place model of object positions and then using training examples to compute the probability that a pair depict the same place. We also generate an estimate of the relative pose of the cameras. Results on a dataset of 40 urban locations show good recognition performance and pose estimation, even for highly disparate views.

Original language	English
Title of host publication	Proceedings of the British Machine Vision Conference (BMVC)
Publication status	Published - Sept 2013

Handle.net

Persistent link

Cite this

@inproceedings{8ae0eb5ae4364f338d180500a53de80e,

title = "Place recognition from disparate views",

abstract = " We use detected objects as basic features in a semantic place recognition system with the aim of allowing recognition when views are disparate. This is achieved by constructing a 2D place model of object positions and then using training examples to compute the probability that a pair depict the same place. We also generate an estimate of the relative pose of the cameras. Results on a dataset of 40 urban locations show good recognition performance and pose estimation, even for highly disparate views. ",

author = "Rob Frampton and Calway, \{Andrew D\}",

year = "2013",

month = sep,

language = "English",

booktitle = "Proceedings of the British Machine Vision Conference (BMVC)",

}

TY - GEN

T1 - Place recognition from disparate views

AU - Frampton, Rob

AU - Calway, Andrew D

PY - 2013/9

Y1 - 2013/9

N2 - We use detected objects as basic features in a semantic place recognition system with the aim of allowing recognition when views are disparate. This is achieved by constructing a 2D place model of object positions and then using training examples to compute the probability that a pair depict the same place. We also generate an estimate of the relative pose of the cameras. Results on a dataset of 40 urban locations show good recognition performance and pose estimation, even for highly disparate views.

AB - We use detected objects as basic features in a semantic place recognition system with the aim of allowing recognition when views are disparate. This is achieved by constructing a 2D place model of object positions and then using training examples to compute the probability that a pair depict the same place. We also generate an estimate of the relative pose of the cameras. Results on a dataset of 40 urban locations show good recognition performance and pose estimation, even for highly disparate views.

M3 - Conference Contribution (Conference Proceeding)

BT - Proceedings of the British Machine Vision Conference (BMVC)

ER -

Place recognition from disparate views

Abstract

Handle.net

Fingerprint

Cite this