Learning memorability preserving subspace for predicting media memorability

Yang Liu; Zhonglei Gu; Tobey H. Ko

Learning memorability preserving subspace for predicting media memorability

Yang Liu, Zhonglei Gu, Tobey H. Ko

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

This paper describes our approach designed for the MediaEval 2018 Predicting Media Memorability Task. First, a subspace learning method called Memorability Preserving Embedding (MPE) is proposed to learn discriminative subspace from the original feature space according to the memorability scores. Then the Support Vector Regressor (SVR) is applied to the learned subspace for memorability prediction. The prediction performance demonstrates that SVR can achieve good performance even in a very low-dimensional subspace, which implies that the subspace learned by the MPE is capable of preserving important memorability information. Moreover, the results indicate that the short-term memorability is more predictable than the long-term memorability. Copyright held by the owner/author(s).

Original language	English
Title of host publication	Working Notes Proceedings of the MediaEval 2018 Workshop
Editors	Martha Larson, Piyush Arora, Claire-Hélène Demarty, Michael Riegler, Benjamin Bischke, Emmanuel Dellandrea, Mathias Lux, Alastair Porter, Gareth J. F. Jones
Publisher	CEUR-WS
Number of pages	3
Publication status	Published - Oct 2018
Event	MediaEval 2018: Multimedia Benchmark Workshop - Sophia Antipolis, France Duration: 29 Oct 2018 → 31 Oct 2018 https://ceur-ws.org/Vol-2283/

Publication series

Name	CEUR Workshop Proceedings
Publisher	CEUR-WS
Volume	2283
ISSN (Print)	1613-0073

Conference

Conference	MediaEval 2018: Multimedia Benchmark Workshop
Country/Territory	France
City	Sophia Antipolis
Period	29/10/18 → 31/10/18
Internet address	https://ceur-ws.org/Vol-2283/

Access to Document

http://ceur-ws.org/Vol-2283/MediaEval_18_paper_33.pdf

Cite this

Liu, Y., Gu, Z., & Ko, T. H. (2018). Learning memorability preserving subspace for predicting media memorability. In M. Larson, P. Arora, C.-H. Demarty, M. Riegler, B. Bischke, E. Dellandrea, M. Lux, A. Porter, & G. J. F. Jones (Eds.), Working Notes Proceedings of the MediaEval 2018 Workshop (CEUR Workshop Proceedings; Vol. 2283). CEUR-WS. http://ceur-ws.org/Vol-2283/MediaEval_18_paper_33.pdf

Liu, Yang ; Gu, Zhonglei ; Ko, Tobey H. / Learning memorability preserving subspace for predicting media memorability. Working Notes Proceedings of the MediaEval 2018 Workshop. editor / Martha Larson ; Piyush Arora ; Claire-Hélène Demarty ; Michael Riegler ; Benjamin Bischke ; Emmanuel Dellandrea ; Mathias Lux ; Alastair Porter ; Gareth J. F. Jones. CEUR-WS, 2018. (CEUR Workshop Proceedings).

@inproceedings{6440e86cbd154ef7bae1d1aa6d63f3a5,

title = "Learning memorability preserving subspace for predicting media memorability",

abstract = "This paper describes our approach designed for the MediaEval 2018 Predicting Media Memorability Task. First, a subspace learning method called Memorability Preserving Embedding (MPE) is proposed to learn discriminative subspace from the original feature space according to the memorability scores. Then the Support Vector Regressor (SVR) is applied to the learned subspace for memorability prediction. The prediction performance demonstrates that SVR can achieve good performance even in a very low-dimensional subspace, which implies that the subspace learned by the MPE is capable of preserving important memorability information. Moreover, the results indicate that the short-term memorability is more predictable than the long-term memorability. Copyright held by the owner/author(s).",

author = "Yang Liu and Zhonglei Gu and Ko, {Tobey H.}",

note = "Funding Information: This work was supported in part by the National Natural Science Foundation of China (NSFC) under Grant 61503317 and in part by the General Research Fund (GRF) from the Research Grant Council (RGC) of Hong Kong SAR under Project HKBU12202417.; MediaEval 2018: Multimedia Benchmark Workshop ; Conference date: 29-10-2018 Through 31-10-2018",

year = "2018",

month = oct,

language = "English",

series = "CEUR Workshop Proceedings",

publisher = "CEUR-WS",

editor = "Martha Larson and Piyush Arora and Claire-H{\'e}l{\`e}ne Demarty and Michael Riegler and Benjamin Bischke and Emmanuel Dellandrea and Mathias Lux and Alastair Porter and Jones, {Gareth J. F.}",

booktitle = "Working Notes Proceedings of the MediaEval 2018 Workshop",

url = "https://ceur-ws.org/Vol-2283/",

}

Liu, Y, Gu, Z & Ko, TH 2018, Learning memorability preserving subspace for predicting media memorability. in M Larson, P Arora, C-H Demarty, M Riegler, B Bischke, E Dellandrea, M Lux, A Porter & GJF Jones (eds), Working Notes Proceedings of the MediaEval 2018 Workshop. CEUR Workshop Proceedings, vol. 2283, CEUR-WS, MediaEval 2018: Multimedia Benchmark Workshop, Sophia Antipolis, France, 29/10/18. <http://ceur-ws.org/Vol-2283/MediaEval_18_paper_33.pdf>

Learning memorability preserving subspace for predicting media memorability. / Liu, Yang; Gu, Zhonglei; Ko, Tobey H.
Working Notes Proceedings of the MediaEval 2018 Workshop. ed. / Martha Larson; Piyush Arora; Claire-Hélène Demarty; Michael Riegler; Benjamin Bischke; Emmanuel Dellandrea; Mathias Lux; Alastair Porter; Gareth J. F. Jones. CEUR-WS, 2018. (CEUR Workshop Proceedings; Vol. 2283).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Learning memorability preserving subspace for predicting media memorability

AU - Liu, Yang

AU - Gu, Zhonglei

AU - Ko, Tobey H.

N1 - Funding Information: This work was supported in part by the National Natural Science Foundation of China (NSFC) under Grant 61503317 and in part by the General Research Fund (GRF) from the Research Grant Council (RGC) of Hong Kong SAR under Project HKBU12202417.

PY - 2018/10

Y1 - 2018/10

N2 - This paper describes our approach designed for the MediaEval 2018 Predicting Media Memorability Task. First, a subspace learning method called Memorability Preserving Embedding (MPE) is proposed to learn discriminative subspace from the original feature space according to the memorability scores. Then the Support Vector Regressor (SVR) is applied to the learned subspace for memorability prediction. The prediction performance demonstrates that SVR can achieve good performance even in a very low-dimensional subspace, which implies that the subspace learned by the MPE is capable of preserving important memorability information. Moreover, the results indicate that the short-term memorability is more predictable than the long-term memorability. Copyright held by the owner/author(s).

AB - This paper describes our approach designed for the MediaEval 2018 Predicting Media Memorability Task. First, a subspace learning method called Memorability Preserving Embedding (MPE) is proposed to learn discriminative subspace from the original feature space according to the memorability scores. Then the Support Vector Regressor (SVR) is applied to the learned subspace for memorability prediction. The prediction performance demonstrates that SVR can achieve good performance even in a very low-dimensional subspace, which implies that the subspace learned by the MPE is capable of preserving important memorability information. Moreover, the results indicate that the short-term memorability is more predictable than the long-term memorability. Copyright held by the owner/author(s).

UR - http://ceur-ws.org/Vol-2283/

UR - http://www.scopus.com/inward/record.url?scp=85059837561&partnerID=8YFLogxK

M3 - Conference proceeding

AN - SCOPUS:85059837561

T3 - CEUR Workshop Proceedings

BT - Working Notes Proceedings of the MediaEval 2018 Workshop

A2 - Larson, Martha

A2 - Arora, Piyush

A2 - Demarty, Claire-Hélène

A2 - Riegler, Michael

A2 - Bischke, Benjamin

A2 - Dellandrea, Emmanuel

A2 - Lux, Mathias

A2 - Porter, Alastair

A2 - Jones, Gareth J. F.

PB - CEUR-WS

T2 - MediaEval 2018: Multimedia Benchmark Workshop

Y2 - 29 October 2018 through 31 October 2018

ER -

Learning memorability preserving subspace for predicting media memorability

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this