Learning geo-temporal image features

Menghua Zhai; Tawfiq Salem; Connor Greenwell; Scott Workman; Nathan Jacobs; Robert Pless

Learning geo-temporal image features

Menghua Zhai, Tawfiq Salem, Connor Greenwell, Scott Workman, Nathan Jacobs, Robert Pless

Research output: Contribution to conference › Paper › peer-review

Abstract

We propose to implicitly learn to extract geo-temporal image features, which are mid-level features related to when and where an image was captured, by explicitly optimizing for a set of location and time estimation tasks. To train our method, we take advantage of a large image dataset, captured by outdoor webcams and cell phones. The only form of supervision we provide are the known capture time and location of each image. We find that our approach learns features that are related to natural appearance changes in outdoor scenes. Additionally, we demonstrate the application of these geo-temporal features to time and location estimation.

Original language	English
State	Published - 2019
Event	29th British Machine Vision Conference, BMVC 2018 - Newcastle, United Kingdom Duration: Sep 3 2018 → Sep 6 2018

Conference

Conference	29th British Machine Vision Conference, BMVC 2018
Country/Territory	United Kingdom
City	Newcastle
Period	09/3/18 → 09/6/18

Cite this

@conference{123d4d3366b54386b60b813148415bfa,

title = "Learning geo-temporal image features",

abstract = "We propose to implicitly learn to extract geo-temporal image features, which are mid-level features related to when and where an image was captured, by explicitly optimizing for a set of location and time estimation tasks. To train our method, we take advantage of a large image dataset, captured by outdoor webcams and cell phones. The only form of supervision we provide are the known capture time and location of each image. We find that our approach learns features that are related to natural appearance changes in outdoor scenes. Additionally, we demonstrate the application of these geo-temporal features to time and location estimation.",

author = "Menghua Zhai and Tawfiq Salem and Connor Greenwell and Scott Workman and Nathan Jacobs and Robert Pless",

note = "Publisher Copyright: {\textcopyright} 2018. The copyright of this document resides with its authors.; 29th British Machine Vision Conference, BMVC 2018 ; Conference date: 03-09-2018 Through 06-09-2018",

year = "2019",

language = "English",

}

TY - CONF

T1 - Learning geo-temporal image features

AU - Zhai, Menghua

AU - Salem, Tawfiq

AU - Greenwell, Connor

AU - Workman, Scott

AU - Jacobs, Nathan

AU - Pless, Robert

PY - 2019

Y1 - 2019

N2 - We propose to implicitly learn to extract geo-temporal image features, which are mid-level features related to when and where an image was captured, by explicitly optimizing for a set of location and time estimation tasks. To train our method, we take advantage of a large image dataset, captured by outdoor webcams and cell phones. The only form of supervision we provide are the known capture time and location of each image. We find that our approach learns features that are related to natural appearance changes in outdoor scenes. Additionally, we demonstrate the application of these geo-temporal features to time and location estimation.

AB - We propose to implicitly learn to extract geo-temporal image features, which are mid-level features related to when and where an image was captured, by explicitly optimizing for a set of location and time estimation tasks. To train our method, we take advantage of a large image dataset, captured by outdoor webcams and cell phones. The only form of supervision we provide are the known capture time and location of each image. We find that our approach learns features that are related to natural appearance changes in outdoor scenes. Additionally, we demonstrate the application of these geo-temporal features to time and location estimation.

UR - http://www.scopus.com/inward/record.url?scp=85084014717&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85084014717

T2 - 29th British Machine Vision Conference, BMVC 2018

Y2 - 3 September 2018 through 6 September 2018

ER -

Learning geo-temporal image features

Abstract

Conference

Other files and links

Fingerprint

Cite this