Hierarchical modelling and adaptive clustering for real-time summarization of rush videos

Jinchang Ren, J. Jiang

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

In this paper, we provide detailed descriptions of a proposed new algorithm for video summarization, which are also included in our submission to TRECVID'08 on BBC rush summarization. Firstly, rush videos are hierarchically modeled using the formal language technique. Secondly, shot detections are applied to introduce a new concept of V-unit for structuring videos in line with the hierarchical model, and thus junk frames within the model are effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to determine retakes for redundancy removal. Finally, each most representative shot selected from every cluster is ranked according to its length and sum of activity level for summarization. Competitive results have been achieved to prove the effectiveness and efficiency of our techniques, which are fully implemented in the compressed domain. Our work does not require high-level semantics such as object detection and speech/audio analysis which provides a more flexible and general solution for this topic.
LanguageEnglish
Pages906-917
Number of pages12
JournalIEEE Transactions on Multimedia
Volume11
Issue number5
DOIs
Publication statusPublished - Aug 2009

Fingerprint

Formal languages
Redundancy
Semantics
Object detection

Keywords

  • multimedia
  • pattern clustering
  • object detection
  • adaptive clustering
  • hierarchical modelling

Cite this

@article{61ba95c067b545aa8ce634a9370d64c1,
title = "Hierarchical modelling and adaptive clustering for real-time summarization of rush videos",
abstract = "In this paper, we provide detailed descriptions of a proposed new algorithm for video summarization, which are also included in our submission to TRECVID'08 on BBC rush summarization. Firstly, rush videos are hierarchically modeled using the formal language technique. Secondly, shot detections are applied to introduce a new concept of V-unit for structuring videos in line with the hierarchical model, and thus junk frames within the model are effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to determine retakes for redundancy removal. Finally, each most representative shot selected from every cluster is ranked according to its length and sum of activity level for summarization. Competitive results have been achieved to prove the effectiveness and efficiency of our techniques, which are fully implemented in the compressed domain. Our work does not require high-level semantics such as object detection and speech/audio analysis which provides a more flexible and general solution for this topic.",
keywords = "multimedia, pattern clustering, object detection, adaptive clustering, hierarchical modelling",
author = "Jinchang Ren and J. Jiang",
year = "2009",
month = "8",
doi = "10.1109/TMM.2009.2021782",
language = "English",
volume = "11",
pages = "906--917",
journal = "IEEE Transactions on Multimedia",
issn = "1520-9210",
number = "5",

}

Hierarchical modelling and adaptive clustering for real-time summarization of rush videos. / Ren, Jinchang; Jiang , J.

In: IEEE Transactions on Multimedia, Vol. 11, No. 5, 08.2009, p. 906-917.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hierarchical modelling and adaptive clustering for real-time summarization of rush videos

AU - Ren, Jinchang

AU - Jiang , J.

PY - 2009/8

Y1 - 2009/8

N2 - In this paper, we provide detailed descriptions of a proposed new algorithm for video summarization, which are also included in our submission to TRECVID'08 on BBC rush summarization. Firstly, rush videos are hierarchically modeled using the formal language technique. Secondly, shot detections are applied to introduce a new concept of V-unit for structuring videos in line with the hierarchical model, and thus junk frames within the model are effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to determine retakes for redundancy removal. Finally, each most representative shot selected from every cluster is ranked according to its length and sum of activity level for summarization. Competitive results have been achieved to prove the effectiveness and efficiency of our techniques, which are fully implemented in the compressed domain. Our work does not require high-level semantics such as object detection and speech/audio analysis which provides a more flexible and general solution for this topic.

AB - In this paper, we provide detailed descriptions of a proposed new algorithm for video summarization, which are also included in our submission to TRECVID'08 on BBC rush summarization. Firstly, rush videos are hierarchically modeled using the formal language technique. Secondly, shot detections are applied to introduce a new concept of V-unit for structuring videos in line with the hierarchical model, and thus junk frames within the model are effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to determine retakes for redundancy removal. Finally, each most representative shot selected from every cluster is ranked according to its length and sum of activity level for summarization. Competitive results have been achieved to prove the effectiveness and efficiency of our techniques, which are fully implemented in the compressed domain. Our work does not require high-level semantics such as object detection and speech/audio analysis which provides a more flexible and general solution for this topic.

KW - multimedia

KW - pattern clustering

KW - object detection

KW - adaptive clustering

KW - hierarchical modelling

UR - http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6046

U2 - 10.1109/TMM.2009.2021782

DO - 10.1109/TMM.2009.2021782

M3 - Article

VL - 11

SP - 906

EP - 917

JO - IEEE Transactions on Multimedia

T2 - IEEE Transactions on Multimedia

JF - IEEE Transactions on Multimedia

SN - 1520-9210

IS - 5

ER -