Intrinsic Image Popularity Assessment

Keyan Ding; Kede Ma; Shiqi Wang

doi:10.1145/3343031.3351007

Intrinsic Image Popularity Assessment

Keyan Ding
, Kede Ma
, Shiqi Wang^*

^*Corresponding author for this work

Department of Computer Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

31 Citations (Scopus)

Abstract

The goal of research in automatic image popularity assessment (IPA) is to develop computational models that can accurately predict the potential of a social image to go viral on the Internet. Here, we aim to single out the contribution of visual content to image popularity, i.e., intrinsic image popularity. Specifically, we first describe a probabilistic method to generate massive popularity-discriminable image pairs, based on which the first large-scale image database for intrinsic IPA (I²PA) is established. We then develop computational models for I²PA based on deep neural networks, optimizing for ranking consistency with millions of popularity-discriminable image pairs. Experiments on Instagram and other social platforms demonstrate that the optimized model performs favorably against existing methods, exhibits reasonable generalizability on different databases, and even surpasses human-level performance on Instagram. In addition, we conduct a psychophysical experiment to analyze various aspects of human behavior in I²PA.

Original language	English
Title of host publication	MM’19
Subtitle of host publication	Proceedings of the 27th ACM International Conference on Multimedia
Place of Publication	New York
Publisher	Association for Computing Machinery
Pages	1979-1987
ISBN (Electronic)	9781450368896
DOIs	https://doi.org/10.1145/3343031.3351007
Publication status	Published - Oct 2019
Event	27th ACM International Conference on Multimedia (MM '19) - NICE ACROPOLIS Convention Center, Nice, France Duration: 21 Oct 2019 → 25 Oct 2019

Publication series

Name	MM - Proceedings of the ACM International Conference on Multimedia

Conference

Conference	27th ACM International Conference on Multimedia (MM '19)
Abbreviated title	MM 2019
Place	France
City	Nice
Period	21/10/19 → 25/10/19

Bibliographical note

Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

Research Keywords

Deep neural networks
Human behavior analysis
Intrinsic image popularity
Learning-to-rank

Access Output

10.1145/3343031.3351007

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{065d53e39cf4456d9dccca79ac15c267,

title = "Intrinsic Image Popularity Assessment",

abstract = "The goal of research in automatic image popularity assessment (IPA) is to develop computational models that can accurately predict the potential of a social image to go viral on the Internet. Here, we aim to single out the contribution of visual content to image popularity, i.e., intrinsic image popularity. Specifically, we first describe a probabilistic method to generate massive popularity-discriminable image pairs, based on which the first large-scale image database for intrinsic IPA (I2PA) is established. We then develop computational models for I2PA based on deep neural networks, optimizing for ranking consistency with millions of popularity-discriminable image pairs. Experiments on Instagram and other social platforms demonstrate that the optimized model performs favorably against existing methods, exhibits reasonable generalizability on different databases, and even surpasses human-level performance on Instagram. In addition, we conduct a psychophysical experiment to analyze various aspects of human behavior in I2PA.",

keywords = "Deep neural networks, Human behavior analysis, Intrinsic image popularity, Learning-to-rank",

author = "Keyan Ding and Kede Ma and Shiqi Wang",

note = "Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).; 27th ACM International Conference on Multimedia (MM '19), MM 2019 ; Conference date: 21-10-2019 Through 25-10-2019",

year = "2019",

month = oct,

doi = "10.1145/3343031.3351007",

language = "English",

series = "MM - Proceedings of the ACM International Conference on Multimedia",

publisher = "Association for Computing Machinery",

pages = "1979--1987",

booktitle = "MM{\textquoteright}19",

address = "United States",

}

Ding, K , Ma, K & Wang, S 2019, Intrinsic Image Popularity Assessment. in MM’19: Proceedings of the 27th ACM International Conference on Multimedia. MM - Proceedings of the ACM International Conference on Multimedia, Association for Computing Machinery, New York, pp. 1979-1987, 27th ACM International Conference on Multimedia (MM '19), Nice, France, 21/10/19. https://doi.org/10.1145/3343031.3351007

Intrinsic Image Popularity Assessment. / Ding, Keyan ; Ma, Kede ; Wang, Shiqi.
MM’19: Proceedings of the 27th ACM International Conference on Multimedia. New York: Association for Computing Machinery, 2019. p. 1979-1987 (MM - Proceedings of the ACM International Conference on Multimedia).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Intrinsic Image Popularity Assessment

AU - Ding, Keyan

AU - Ma, Kede

AU - Wang, Shiqi

N1 - Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

PY - 2019/10

Y1 - 2019/10

N2 - The goal of research in automatic image popularity assessment (IPA) is to develop computational models that can accurately predict the potential of a social image to go viral on the Internet. Here, we aim to single out the contribution of visual content to image popularity, i.e., intrinsic image popularity. Specifically, we first describe a probabilistic method to generate massive popularity-discriminable image pairs, based on which the first large-scale image database for intrinsic IPA (I2PA) is established. We then develop computational models for I2PA based on deep neural networks, optimizing for ranking consistency with millions of popularity-discriminable image pairs. Experiments on Instagram and other social platforms demonstrate that the optimized model performs favorably against existing methods, exhibits reasonable generalizability on different databases, and even surpasses human-level performance on Instagram. In addition, we conduct a psychophysical experiment to analyze various aspects of human behavior in I2PA.

AB - The goal of research in automatic image popularity assessment (IPA) is to develop computational models that can accurately predict the potential of a social image to go viral on the Internet. Here, we aim to single out the contribution of visual content to image popularity, i.e., intrinsic image popularity. Specifically, we first describe a probabilistic method to generate massive popularity-discriminable image pairs, based on which the first large-scale image database for intrinsic IPA (I2PA) is established. We then develop computational models for I2PA based on deep neural networks, optimizing for ranking consistency with millions of popularity-discriminable image pairs. Experiments on Instagram and other social platforms demonstrate that the optimized model performs favorably against existing methods, exhibits reasonable generalizability on different databases, and even surpasses human-level performance on Instagram. In addition, we conduct a psychophysical experiment to analyze various aspects of human behavior in I2PA.

KW - Deep neural networks

KW - Human behavior analysis

KW - Intrinsic image popularity

KW - Learning-to-rank

UR - https://www.scopus.com/pages/publications/85074808629

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85074808629&origin=recordpage

U2 - 10.1145/3343031.3351007

DO - 10.1145/3343031.3351007

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - MM - Proceedings of the ACM International Conference on Multimedia

SP - 1979

EP - 1987

BT - MM’19

PB - Association for Computing Machinery

CY - New York

T2 - 27th ACM International Conference on Multimedia (MM '19)

Y2 - 21 October 2019 through 25 October 2019

ER -

Intrinsic Image Popularity Assessment

Abstract

Publication series

Conference

Bibliographical note

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Deep Learning-Based Image Quality and Popularity Assessment

Cite this

Intrinsic Image Popularity Assessment

Abstract

Publication series

Conference

Bibliographical note

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Student theses

Deep Learning-Based Image Quality and Popularity Assessment

Cite this