Estimating the camera direction of a geotagged image using reference images

Minwoo Park; Jiebo Luo; Robert T. Collins; Yanxi Liu

doi:10.1016/j.patcog.2014.03.002

Estimating the camera direction of a geotagged image using reference images

Minwoo Park
, Jiebo Luo
, Robert T. Collins
, Yanxi Liu

Research output: Contribution to journal › Article › peer-review

14 Scopus citations

Abstract

Millions of smart phones and GPS-equipped digital cameras sold each year, as well as photo-sharing websites such as Picasa and Panoramio have enabled personal photos to be associated with geographic information. It has been shown by recent research results that the additional global positioning system (GPS) information helps visual recognition for geotagged photos by providing valuable location context. However, the current GPS data only identifies the camera location, leaving the camera viewing direction uncertain within the possible scope of 360°. To produce more precise photo location information, i.e. the viewing direction for geotagged photos, we utilize both Google Street View and Google Earth satellite images. Our proposed system is two-pronged: (1) visual matching between a user photo and any available street views in the vicinity can determine the viewing direction, and (2) near-orthogonal view matching between a user photo taken on the ground and the overhead satellite view at the user geo-location can compute the viewing direction when only the satellite view is available. Experimental results have shown the effectiveness of the proposed framework.

Original language	English (US)
Pages (from-to)	2880-2893
Number of pages	14
Journal	Pattern Recognition
Volume	47
Issue number	9
DOIs	https://doi.org/10.1016/j.patcog.2014.03.002
State	Published - Sep 2014

All Science Journal Classification (ASJC) codes

Software
Signal Processing
Computer Vision and Pattern Recognition
Artificial Intelligence

Access to Document

10.1016/j.patcog.2014.03.002

Cite this

@article{2e7b7b1058fc437b9d38c47d695cce10,

title = "Estimating the camera direction of a geotagged image using reference images",

abstract = "Millions of smart phones and GPS-equipped digital cameras sold each year, as well as photo-sharing websites such as Picasa and Panoramio have enabled personal photos to be associated with geographic information. It has been shown by recent research results that the additional global positioning system (GPS) information helps visual recognition for geotagged photos by providing valuable location context. However, the current GPS data only identifies the camera location, leaving the camera viewing direction uncertain within the possible scope of 360°. To produce more precise photo location information, i.e. the viewing direction for geotagged photos, we utilize both Google Street View and Google Earth satellite images. Our proposed system is two-pronged: (1) visual matching between a user photo and any available street views in the vicinity can determine the viewing direction, and (2) near-orthogonal view matching between a user photo taken on the ground and the overhead satellite view at the user geo-location can compute the viewing direction when only the satellite view is available. Experimental results have shown the effectiveness of the proposed framework.",

author = "Minwoo Park and Jiebo Luo and Collins, \{Robert T.\} and Yanxi Liu",

note = "Funding Information: Yanxi Liu received the B.S. degree in physics/electrical engineering from Beijing, China, the Ph.D. degree in computer science for group theory applications in robotics from the University of Massachusetts, Amherst, and postdoctoral training at LIFIA/IMAG, Grenoble, France. She also spent one year at the US National Science Foundation (NSF) Center for Discrete Mathematics and Theoretical Computer Science (DIMACS) with an NSF Research Education Fellowship Award, and a sabbatical-semester in the radiology department of UPMC. She serves as the co-director of the Lab for Perception, Action, and Cognition (LPAC) and is a tenured faculty member of the Computer Science Engineering and Electrical Engineering Departments of The Pennsylvania State University. Before joining PSU, she was an associate research professor in the Robotics Institute of Carnegie Mellon University. Her research interests span a wide range of applications, including computer vision, computer graphics, robotics, human perception, and computer-aided diagnosis in medicine, with a theme on computational symmetry/regularity and discriminative subspace learning from large, multimedia datasets. She co-chaired the International Workshop on “Computer Vision in Biomedical Image Applications” (ICCV 2005), and is chairing the first US NSF funded international competition on ”Symmetry Detection from Real World Images” (CVPR 2011). She is a senior member of the IEEE and the IEEE Computer Society.",

year = "2014",

month = sep,

doi = "10.1016/j.patcog.2014.03.002",

language = "English (US)",

volume = "47",

pages = "2880--2893",

journal = "Pattern Recognition",

issn = "0031-3203",

publisher = "Elsevier Ltd",

number = "9",

}

TY - JOUR

T1 - Estimating the camera direction of a geotagged image using reference images

AU - Park, Minwoo

AU - Luo, Jiebo

AU - Collins, Robert T.

AU - Liu, Yanxi

N1 - Funding Information: Yanxi Liu received the B.S. degree in physics/electrical engineering from Beijing, China, the Ph.D. degree in computer science for group theory applications in robotics from the University of Massachusetts, Amherst, and postdoctoral training at LIFIA/IMAG, Grenoble, France. She also spent one year at the US National Science Foundation (NSF) Center for Discrete Mathematics and Theoretical Computer Science (DIMACS) with an NSF Research Education Fellowship Award, and a sabbatical-semester in the radiology department of UPMC. She serves as the co-director of the Lab for Perception, Action, and Cognition (LPAC) and is a tenured faculty member of the Computer Science Engineering and Electrical Engineering Departments of The Pennsylvania State University. Before joining PSU, she was an associate research professor in the Robotics Institute of Carnegie Mellon University. Her research interests span a wide range of applications, including computer vision, computer graphics, robotics, human perception, and computer-aided diagnosis in medicine, with a theme on computational symmetry/regularity and discriminative subspace learning from large, multimedia datasets. She co-chaired the International Workshop on “Computer Vision in Biomedical Image Applications” (ICCV 2005), and is chairing the first US NSF funded international competition on ”Symmetry Detection from Real World Images” (CVPR 2011). She is a senior member of the IEEE and the IEEE Computer Society.

PY - 2014/9

Y1 - 2014/9

N2 - Millions of smart phones and GPS-equipped digital cameras sold each year, as well as photo-sharing websites such as Picasa and Panoramio have enabled personal photos to be associated with geographic information. It has been shown by recent research results that the additional global positioning system (GPS) information helps visual recognition for geotagged photos by providing valuable location context. However, the current GPS data only identifies the camera location, leaving the camera viewing direction uncertain within the possible scope of 360°. To produce more precise photo location information, i.e. the viewing direction for geotagged photos, we utilize both Google Street View and Google Earth satellite images. Our proposed system is two-pronged: (1) visual matching between a user photo and any available street views in the vicinity can determine the viewing direction, and (2) near-orthogonal view matching between a user photo taken on the ground and the overhead satellite view at the user geo-location can compute the viewing direction when only the satellite view is available. Experimental results have shown the effectiveness of the proposed framework.

AB - Millions of smart phones and GPS-equipped digital cameras sold each year, as well as photo-sharing websites such as Picasa and Panoramio have enabled personal photos to be associated with geographic information. It has been shown by recent research results that the additional global positioning system (GPS) information helps visual recognition for geotagged photos by providing valuable location context. However, the current GPS data only identifies the camera location, leaving the camera viewing direction uncertain within the possible scope of 360°. To produce more precise photo location information, i.e. the viewing direction for geotagged photos, we utilize both Google Street View and Google Earth satellite images. Our proposed system is two-pronged: (1) visual matching between a user photo and any available street views in the vicinity can determine the viewing direction, and (2) near-orthogonal view matching between a user photo taken on the ground and the overhead satellite view at the user geo-location can compute the viewing direction when only the satellite view is available. Experimental results have shown the effectiveness of the proposed framework.

UR - https://www.scopus.com/pages/publications/84900804976

UR - https://www.scopus.com/pages/publications/84900804976#tab=citedBy

U2 - 10.1016/j.patcog.2014.03.002

DO - 10.1016/j.patcog.2014.03.002

M3 - Article

AN - SCOPUS:84900804976

SN - 0031-3203

VL - 47

SP - 2880

EP - 2893

JO - Pattern Recognition

JF - Pattern Recognition

IS - 9

ER -

Estimating the camera direction of a geotagged image using reference images

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this