Attention estimation by simultaneous analysis of viewer and view

Ashish Tawari; Andreas Møgelmose; Sujitha Martin; Thomas B. Moeslund; Mohan M. Trivedi

doi:10.1109/ITSC.2014.6957880

Attention estimation by simultaneous analysis of viewer and view

Ashish Tawari, Andreas Møgelmose, Sujitha Martin, Thomas B. Moeslund, Mohan M. Trivedi

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

24 Citations (Scopus)

535 Downloads (Pure)

Abstract

This paper introduces a system for estimating the attention of a driver wearing a first person view camera using salient objects to improve gaze estimation. A challenging data set of pedestrians crossing intersections has been captured using Google Glass worn by a driver. A challenge unique to first person view from cars is that the interior of the car can take up a large part of the image. The proposed system automatically filters out the dashboard of the car, along with other parts of the instrumentation. The remaining area is used as a region of interest for a pedestrian detector. Two cameras looking at the driver are used to determine the direction of the driver’s gaze, by examining the eye corners and the center of the iris. This coarse gaze estimation is then linked to the detected pedestrians to determine which pedestrian the driver is focused on at any given time.

Original language	English
Title of host publication	IEEE 17th International Conference on Intelligent Transportation Systems (ITSC), 2014
Number of pages	7
Publisher	IEEE Press
Publication date	8 Oct 2014
Pages	1381-1387
ISBN (Print)	9781479960798
DOIs	https://doi.org/10.1109/ITSC.2014.6957880
Publication status	Published - 8 Oct 2014
Event	International Conference on Intelligent Transportation Systems, 2014 - Qingdao, China Duration: 8 Oct 2014 → 11 Oct 2014 Conference number: 17

Conference

Conference	International Conference on Intelligent Transportation Systems, 2014
Number	17
Country/Territory	China
City	Qingdao
Period	08/10/2014 → 11/10/2014

Access to Document

10.1109/ITSC.2014.6957880

Attention estimation by simultaneous analysis of viewer and viewAccepted author manuscript, 1.8 MB

http://www.proceedings.com/24373.html

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{11645c9b257d465790a09c839f720462,

title = "Attention estimation by simultaneous analysis of viewer and view",

abstract = "This paper introduces a system for estimating the attention of a driver wearing a first person view camera using salient objects to improve gaze estimation. A challenging data set of pedestrians crossing intersections has been captured using Google Glass worn by a driver. A challenge unique to first person view from cars is that the interior of the car can take up a large part of the image. The proposed system automatically filters out the dashboard of the car, along with other parts of the instrumentation. The remaining area is used as a region of interest for a pedestrian detector. Two cameras looking at the driver are used to determine the direction of the driver{\textquoteright}s gaze, by examining the eye corners and the center of the iris. This coarse gaze estimation is then linked to the detected pedestrians to determine which pedestrian the driver is focused on at any given time.",

author = "Ashish Tawari and Andreas M{\o}gelmose and Sujitha Martin and Moeslund, {Thomas B.} and Trivedi, {Mohan M.}",

year = "2014",

month = oct,

day = "8",

doi = "10.1109/ITSC.2014.6957880",

language = "English",

isbn = "9781479960798",

pages = "1381--1387",

booktitle = "IEEE 17th International Conference on Intelligent Transportation Systems (ITSC), 2014",

publisher = "IEEE Press",

note = "International Conference on Intelligent Transportation Systems, 2014 ; Conference date: 08-10-2014 Through 11-10-2014",

}

Tawari, A, Møgelmose, A, Martin, S, Moeslund, TB & Trivedi, MM 2014, Attention estimation by simultaneous analysis of viewer and view. in IEEE 17th International Conference on Intelligent Transportation Systems (ITSC), 2014. IEEE Press, pp. 1381-1387, International Conference on Intelligent Transportation Systems, 2014, Qingdao, China, 08/10/2014. https://doi.org/10.1109/ITSC.2014.6957880

Attention estimation by simultaneous analysis of viewer and view. / Tawari, Ashish; Møgelmose, Andreas; Martin, Sujitha et al.
IEEE 17th International Conference on Intelligent Transportation Systems (ITSC), 2014. IEEE Press, 2014. p. 1381-1387.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Attention estimation by simultaneous analysis of viewer and view

AU - Tawari, Ashish

AU - Møgelmose, Andreas

AU - Martin, Sujitha

AU - Moeslund, Thomas B.

AU - Trivedi, Mohan M.

N1 - Conference code: 17

PY - 2014/10/8

Y1 - 2014/10/8

N2 - This paper introduces a system for estimating the attention of a driver wearing a first person view camera using salient objects to improve gaze estimation. A challenging data set of pedestrians crossing intersections has been captured using Google Glass worn by a driver. A challenge unique to first person view from cars is that the interior of the car can take up a large part of the image. The proposed system automatically filters out the dashboard of the car, along with other parts of the instrumentation. The remaining area is used as a region of interest for a pedestrian detector. Two cameras looking at the driver are used to determine the direction of the driver’s gaze, by examining the eye corners and the center of the iris. This coarse gaze estimation is then linked to the detected pedestrians to determine which pedestrian the driver is focused on at any given time.

AB - This paper introduces a system for estimating the attention of a driver wearing a first person view camera using salient objects to improve gaze estimation. A challenging data set of pedestrians crossing intersections has been captured using Google Glass worn by a driver. A challenge unique to first person view from cars is that the interior of the car can take up a large part of the image. The proposed system automatically filters out the dashboard of the car, along with other parts of the instrumentation. The remaining area is used as a region of interest for a pedestrian detector. Two cameras looking at the driver are used to determine the direction of the driver’s gaze, by examining the eye corners and the center of the iris. This coarse gaze estimation is then linked to the detected pedestrians to determine which pedestrian the driver is focused on at any given time.

U2 - 10.1109/ITSC.2014.6957880

DO - 10.1109/ITSC.2014.6957880

M3 - Article in proceeding

SN - 9781479960798

SP - 1381

EP - 1387

BT - IEEE 17th International Conference on Intelligent Transportation Systems (ITSC), 2014

PB - IEEE Press

T2 - International Conference on Intelligent Transportation Systems, 2014

Y2 - 8 October 2014 through 11 October 2014

ER -

Attention estimation by simultaneous analysis of viewer and view

Abstract

Conference

Access to Document

AUB Link

Fingerprint

Cite this