PET: An eye-tracking dataset for animal-centric Pascal object classes

Syed Omer Gilani; Ramanathan Subramanian; Yan Yan; David Melcher; Nicu Sebe; Stefan Winkler

doi:10.1109/ICME.2015.7177450

PET: An eye-tracking dataset for animal-centric Pascal object classes

Syed Omer Gilani, Ramanathan Subramanian, Yan Yan, David Melcher, Nicu Sebe, Stefan Winkler

Research output: Contribution to conference (non-published works) › Paper › peer-review

13 Citations (Scopus)

Abstract

We present PET- the Pascal animal classes Eye Tracking database. Our database comprises eye movement recordings compiled from forty users for the bird, cat, cow, dog, horse and sheep trainval sets from the VOC 2012 image set. Different from recent eye-tracking databases such as [1, 2], a salient aspect of PET is that it contains eye movements recorded for both the free-viewing and visual search task conditions. While some differences in terms of overall gaze behavior and scanning patterns are observed between the two conditions, a very similar number of fixations are observed on target objects for both conditions. As a utility application, we show how feature pooling around fixated locations enables enhanced (animal) object classification accuracy.

Original language	English
Pages	1-6
Number of pages	6
DOIs	https://doi.org/10.1109/ICME.2015.7177450
Publication status	Published - 4 Aug 2015
Externally published	Yes
Event	IEEE International Conference on Multimedia and Expo, ICME 2015 - Turin, Italy Duration: 29 Jun 2015 → 3 Jul 2015

Conference

Conference	IEEE International Conference on Multimedia and Expo, ICME 2015
Country/Territory	Italy
City	Turin
Period	29/06/15 → 3/07/15

Access to Document

10.1109/ICME.2015.7177450

https://arxiv.org/pdf/1604.01574.pdf

Cite this

@conference{e13776a118c24034b8cea98e1d92381d,

title = "PET: An eye-tracking dataset for animal-centric Pascal object classes",

abstract = "We present PET- the Pascal animal classes Eye Tracking database. Our database comprises eye movement recordings compiled from forty users for the bird, cat, cow, dog, horse and sheep trainval sets from the VOC 2012 image set. Different from recent eye-tracking databases such as [1, 2], a salient aspect of PET is that it contains eye movements recorded for both the free-viewing and visual search task conditions. While some differences in terms of overall gaze behavior and scanning patterns are observed between the two conditions, a very similar number of fixations are observed on target objects for both conditions. As a utility application, we show how feature pooling around fixated locations enables enhanced (animal) object classification accuracy.",

keywords = "Animal-centric object classes, Eye movements, Free-viewing vs. Visual search, Pascal VOC, PET",

author = "Gilani, {Syed Omer} and Ramanathan Subramanian and Yan Yan and David Melcher and Nicu Sebe and Stefan Winkler",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.; IEEE International Conference on Multimedia and Expo, ICME 2015 ; Conference date: 29-06-2015 Through 03-07-2015",

year = "2015",

month = aug,

day = "4",

doi = "10.1109/ICME.2015.7177450",

language = "English",

pages = "1--6",

}

TY - CONF

T1 - PET

T2 - IEEE International Conference on Multimedia and Expo, ICME 2015

AU - Gilani, Syed Omer

AU - Subramanian, Ramanathan

AU - Yan, Yan

AU - Melcher, David

AU - Sebe, Nicu

AU - Winkler, Stefan

PY - 2015/8/4

Y1 - 2015/8/4

N2 - We present PET- the Pascal animal classes Eye Tracking database. Our database comprises eye movement recordings compiled from forty users for the bird, cat, cow, dog, horse and sheep trainval sets from the VOC 2012 image set. Different from recent eye-tracking databases such as [1, 2], a salient aspect of PET is that it contains eye movements recorded for both the free-viewing and visual search task conditions. While some differences in terms of overall gaze behavior and scanning patterns are observed between the two conditions, a very similar number of fixations are observed on target objects for both conditions. As a utility application, we show how feature pooling around fixated locations enables enhanced (animal) object classification accuracy.

AB - We present PET- the Pascal animal classes Eye Tracking database. Our database comprises eye movement recordings compiled from forty users for the bird, cat, cow, dog, horse and sheep trainval sets from the VOC 2012 image set. Different from recent eye-tracking databases such as [1, 2], a salient aspect of PET is that it contains eye movements recorded for both the free-viewing and visual search task conditions. While some differences in terms of overall gaze behavior and scanning patterns are observed between the two conditions, a very similar number of fixations are observed on target objects for both conditions. As a utility application, we show how feature pooling around fixated locations enables enhanced (animal) object classification accuracy.

KW - Animal-centric object classes

KW - Eye movements

KW - Free-viewing vs. Visual search

KW - Pascal VOC

KW - PET

UR - http://www.scopus.com/inward/record.url?scp=84946028641&partnerID=8YFLogxK

U2 - 10.1109/ICME.2015.7177450

DO - 10.1109/ICME.2015.7177450

M3 - Paper

AN - SCOPUS:84946028641

SP - 1

EP - 6

Y2 - 29 June 2015 through 3 July 2015

ER -

PET: An eye-tracking dataset for animal-centric Pascal object classes

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this