Labeling images by interpretation from natural viewing

Karen Guo, Danielle N. Pratt, Angus MacDonald, Paul R Schrater

Research output: Contribution to journalConference article

Abstract

In this paper, we would like to discuss the connection between visual processing and the understanding of an image. While the information of image viewing can be obtained from subjects' eye fixation, the understanding of an image can be obtained from the subjects' description of the given image. Furthermore, we proposed a new image labeling method based on the connection between eye fixation and image description by humans. By generating this new kind of labeling method, we can construct an image dataset with labels that are closer to how humans understand the incoming image. In addition, we would like to discuss the proof that the proposed labels better describe the image compared to other types of labeling systems. Research about the relationship between images and human descriptions can be applied to several different applications. For instance, by analyzing the pairwise similarity of user descriptions, we could have a measurement of the complexity of image content. Another possible application is to use this dataset as a criterion to find the difference in visual processing of individuals with or without certain psychological characteristic.

Original languageEnglish (US)
JournalCEUR Workshop Proceedings
Volume2068
StatePublished - Jan 1 2018
Event2018 Joint ACM IUI Workshops, ACMIUI-WS 2018 - Tokyo, Japan
Duration: Mar 11 2018 → …

Fingerprint

Labeling
Labels
Processing

Keywords

  • Computer vision
  • Eye fixation
  • Image annotation
  • Image representation
  • Scene analysis
  • Vision and scene understanding
  • Visual attention

Cite this

Labeling images by interpretation from natural viewing. / Guo, Karen; Pratt, Danielle N.; MacDonald, Angus; Schrater, Paul R.

In: CEUR Workshop Proceedings, Vol. 2068, 01.01.2018.

Research output: Contribution to journalConference article

@article{66e46e93066a4a7fbcfba580c9b4acf1,
title = "Labeling images by interpretation from natural viewing",
abstract = "In this paper, we would like to discuss the connection between visual processing and the understanding of an image. While the information of image viewing can be obtained from subjects' eye fixation, the understanding of an image can be obtained from the subjects' description of the given image. Furthermore, we proposed a new image labeling method based on the connection between eye fixation and image description by humans. By generating this new kind of labeling method, we can construct an image dataset with labels that are closer to how humans understand the incoming image. In addition, we would like to discuss the proof that the proposed labels better describe the image compared to other types of labeling systems. Research about the relationship between images and human descriptions can be applied to several different applications. For instance, by analyzing the pairwise similarity of user descriptions, we could have a measurement of the complexity of image content. Another possible application is to use this dataset as a criterion to find the difference in visual processing of individuals with or without certain psychological characteristic.",
keywords = "Computer vision, Eye fixation, Image annotation, Image representation, Scene analysis, Vision and scene understanding, Visual attention",
author = "Karen Guo and Pratt, {Danielle N.} and Angus MacDonald and Schrater, {Paul R}",
year = "2018",
month = "1",
day = "1",
language = "English (US)",
volume = "2068",
journal = "CEUR Workshop Proceedings",
issn = "1613-0073",
publisher = "CEUR-WS",

}

TY - JOUR

T1 - Labeling images by interpretation from natural viewing

AU - Guo, Karen

AU - Pratt, Danielle N.

AU - MacDonald, Angus

AU - Schrater, Paul R

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In this paper, we would like to discuss the connection between visual processing and the understanding of an image. While the information of image viewing can be obtained from subjects' eye fixation, the understanding of an image can be obtained from the subjects' description of the given image. Furthermore, we proposed a new image labeling method based on the connection between eye fixation and image description by humans. By generating this new kind of labeling method, we can construct an image dataset with labels that are closer to how humans understand the incoming image. In addition, we would like to discuss the proof that the proposed labels better describe the image compared to other types of labeling systems. Research about the relationship between images and human descriptions can be applied to several different applications. For instance, by analyzing the pairwise similarity of user descriptions, we could have a measurement of the complexity of image content. Another possible application is to use this dataset as a criterion to find the difference in visual processing of individuals with or without certain psychological characteristic.

AB - In this paper, we would like to discuss the connection between visual processing and the understanding of an image. While the information of image viewing can be obtained from subjects' eye fixation, the understanding of an image can be obtained from the subjects' description of the given image. Furthermore, we proposed a new image labeling method based on the connection between eye fixation and image description by humans. By generating this new kind of labeling method, we can construct an image dataset with labels that are closer to how humans understand the incoming image. In addition, we would like to discuss the proof that the proposed labels better describe the image compared to other types of labeling systems. Research about the relationship between images and human descriptions can be applied to several different applications. For instance, by analyzing the pairwise similarity of user descriptions, we could have a measurement of the complexity of image content. Another possible application is to use this dataset as a criterion to find the difference in visual processing of individuals with or without certain psychological characteristic.

KW - Computer vision

KW - Eye fixation

KW - Image annotation

KW - Image representation

KW - Scene analysis

KW - Vision and scene understanding

KW - Visual attention

UR - http://www.scopus.com/inward/record.url?scp=85044525142&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044525142&partnerID=8YFLogxK

M3 - Conference article

VL - 2068

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

SN - 1613-0073

ER -