RoleNet: Treat a movie as a small society

Chung Yi Weng, Wei Ta Chu, Ja Ling Wu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

We present a brave new way to analyze movie content, from the perspectives of the relationships between roles rather than low-level audiovisual features. Interactions between roles in a movie resemble human behaviors in a society. Roles' actions lead the story and make viewers understand what directors want to present. In this paper, we introduce the idea of social network analysis to model the relationships of actors/actresses as a network, called RoleNet. Through analyzing this network, the proposed approach automatically determines the leading roles and the communities embedded in movies. We also describe an implementation framework to realize the proposed model. The experimental results show that the proposed methods can effectively capture social characteristics in movies. It's believed that this idea provides a different way to approach movie understanding.

Original languageEnglish
Title of host publicationInternational Multimedia Conference, MM'07 - Proceedings of the 9th ACM SIG Multimedia International Workshop on Multimedia Information Retrieval, MIR'07
Pages51-60
Number of pages10
DOIs
Publication statusPublished - 2007
EventInternational Multimedia Conference, MM'07 - 9th ACM SIG Multimedia International Workshop on Multimedia Information Retrieval, MIR'07 - Augsburg, Bavaria, Germany
Duration: 2007 Sep 282007 Sep 28

Publication series

NameProceedings of the ACM International Multimedia Conference and Exhibition

Conference

ConferenceInternational Multimedia Conference, MM'07 - 9th ACM SIG Multimedia International Workshop on Multimedia Information Retrieval, MIR'07
CountryGermany
CityAugsburg, Bavaria
Period07-09-2807-09-28

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Fingerprint Dive into the research topics of 'RoleNet: Treat a movie as a small society'. Together they form a unique fingerprint.

Cite this