A content-based approach for detecting highlights in action movies

Mei-Chen Yeh, Yen Wei Tsai, Hao Chen Hsu

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Although detecting highlights in films is a trivial task for humans, previous studies have not determined whether a computer can be equipped with this capability. In this paper, we present a content-based system that automatically detects highlight scenes and predicts highlight scores in action movies. In particular, high-level image attributes and an early event detection approach are applied. Dissimilar to current learning-based approaches that model the relationship between the whole highlight and corresponding audiovisual features, the proposed system studies the temporal changes of a set of general features from a nonhighlight to a highlight scene. The experimental results indicate that achieving the highlight detection task is technically feasible. It also provides critical insights into understanding the feasibility of solving this challenging problem. For example, both audio and visual features are crucial and the filming style can be captured using high-level image attributes, which further improve the overall detection performance.

Original languageEnglish
Pages (from-to)287-295
Number of pages9
JournalMultimedia Systems
Volume22
Issue number3
DOIs
Publication statusPublished - 2016 Jan 1

Keywords

  • Affective computing
  • Cross-modality integration
  • Highlight detection
  • Media content analysis

ASJC Scopus subject areas

  • Software
  • Information Systems
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

A content-based approach for detecting highlights in action movies. / Yeh, Mei-Chen; Tsai, Yen Wei; Hsu, Hao Chen.

In: Multimedia Systems, Vol. 22, No. 3, 01.01.2016, p. 287-295.

Research output: Contribution to journalArticle

Yeh, Mei-Chen ; Tsai, Yen Wei ; Hsu, Hao Chen. / A content-based approach for detecting highlights in action movies. In: Multimedia Systems. 2016 ; Vol. 22, No. 3. pp. 287-295.
@article{1a6352ed6e344112b9fb61f677eb1610,
title = "A content-based approach for detecting highlights in action movies",
abstract = "Although detecting highlights in films is a trivial task for humans, previous studies have not determined whether a computer can be equipped with this capability. In this paper, we present a content-based system that automatically detects highlight scenes and predicts highlight scores in action movies. In particular, high-level image attributes and an early event detection approach are applied. Dissimilar to current learning-based approaches that model the relationship between the whole highlight and corresponding audiovisual features, the proposed system studies the temporal changes of a set of general features from a nonhighlight to a highlight scene. The experimental results indicate that achieving the highlight detection task is technically feasible. It also provides critical insights into understanding the feasibility of solving this challenging problem. For example, both audio and visual features are crucial and the filming style can be captured using high-level image attributes, which further improve the overall detection performance.",
keywords = "Affective computing, Cross-modality integration, Highlight detection, Media content analysis",
author = "Mei-Chen Yeh and Tsai, {Yen Wei} and Hsu, {Hao Chen}",
year = "2016",
month = "1",
day = "1",
doi = "10.1007/s00530-015-0457-6",
language = "English",
volume = "22",
pages = "287--295",
journal = "Multimedia Systems",
issn = "0942-4962",
publisher = "Springer Verlag",
number = "3",

}

TY - JOUR

T1 - A content-based approach for detecting highlights in action movies

AU - Yeh, Mei-Chen

AU - Tsai, Yen Wei

AU - Hsu, Hao Chen

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Although detecting highlights in films is a trivial task for humans, previous studies have not determined whether a computer can be equipped with this capability. In this paper, we present a content-based system that automatically detects highlight scenes and predicts highlight scores in action movies. In particular, high-level image attributes and an early event detection approach are applied. Dissimilar to current learning-based approaches that model the relationship between the whole highlight and corresponding audiovisual features, the proposed system studies the temporal changes of a set of general features from a nonhighlight to a highlight scene. The experimental results indicate that achieving the highlight detection task is technically feasible. It also provides critical insights into understanding the feasibility of solving this challenging problem. For example, both audio and visual features are crucial and the filming style can be captured using high-level image attributes, which further improve the overall detection performance.

AB - Although detecting highlights in films is a trivial task for humans, previous studies have not determined whether a computer can be equipped with this capability. In this paper, we present a content-based system that automatically detects highlight scenes and predicts highlight scores in action movies. In particular, high-level image attributes and an early event detection approach are applied. Dissimilar to current learning-based approaches that model the relationship between the whole highlight and corresponding audiovisual features, the proposed system studies the temporal changes of a set of general features from a nonhighlight to a highlight scene. The experimental results indicate that achieving the highlight detection task is technically feasible. It also provides critical insights into understanding the feasibility of solving this challenging problem. For example, both audio and visual features are crucial and the filming style can be captured using high-level image attributes, which further improve the overall detection performance.

KW - Affective computing

KW - Cross-modality integration

KW - Highlight detection

KW - Media content analysis

UR - http://www.scopus.com/inward/record.url?scp=84925426512&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84925426512&partnerID=8YFLogxK

U2 - 10.1007/s00530-015-0457-6

DO - 10.1007/s00530-015-0457-6

M3 - Article

AN - SCOPUS:84925426512

VL - 22

SP - 287

EP - 295

JO - Multimedia Systems

JF - Multimedia Systems

SN - 0942-4962

IS - 3

ER -