Hierarchical modeling and adaptive clustering for real-time summarization of rush videos in TRECVID’08

Jinchang Ren; J. Jiang; Christian Eckes

doi:10.1145/1463563.1463566

Hierarchical modeling and adaptive clustering for real-time summarization of rush videos in TRECVID’08

Jinchang Ren, J. Jiang , Christian Eckes

Electronic And Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

5 Citations (Scopus)

Abstract

In this paper, our techniques used in TRECVID'08 on BBC rush summarization are described. Firstly, rush videos are hierarchical modeled using formal language description. Then, shot detection and V-unit determination are applied for video structuring; junk frames within the model are also effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to remove retakes. Then, each selected shot is ranked according to its length and sum of activity level for summarization. Competitive results have proved the effectiveness and efficiency of our techniques fully implemented in compressed-domain.

Original language	English
Title of host publication	TVS'08 Proc. of the 2nd ACM TRECVID Video Summarisation Workshop
Pages	26-30
Number of pages	5
DOIs	https://doi.org/10.1145/1463563.1463566
Publication status	Published - 2008
Event	TRECVid 2008 - , United Kingdom Duration: 1 Feb 2008 → …

Conference

Conference	TRECVid 2008
Country	United Kingdom
Period	1/02/08 → …

Keywords

adaptive clustering
hierarchical modelling

Access to Document

10.1145/1463563.1463566

Fingerprint Dive into the research topics of 'Hierarchical modeling and adaptive clustering for real-time summarization of rush videos in TRECVID’08'. Together they form a unique fingerprint.

Cite this

@inproceedings{bb4ec7f668ae4c4c87baef04e0f50cc4,

title = "Hierarchical modeling and adaptive clustering for real-time summarization of rush videos in TRECVID{\textquoteright}08",

abstract = "In this paper, our techniques used in TRECVID'08 on BBC rush summarization are described. Firstly, rush videos are hierarchical modeled using formal language description. Then, shot detection and V-unit determination are applied for video structuring; junk frames within the model are also effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to remove retakes. Then, each selected shot is ranked according to its length and sum of activity level for summarization. Competitive results have proved the effectiveness and efficiency of our techniques fully implemented in compressed-domain. ",

keywords = "adaptive clustering, hierarchical modelling",

author = "Jinchang Ren and J. Jiang and Christian Eckes",

year = "2008",

doi = "10.1145/1463563.1463566",

language = "English",

isbn = "978-1-60558-309-9",

pages = "26--30",

booktitle = "TVS'08 Proc. of the 2nd ACM TRECVID Video Summarisation Workshop",

note = "TRECVid 2008 ; Conference date: 01-02-2008",

}

TY - GEN

T1 - Hierarchical modeling and adaptive clustering for real-time summarization of rush videos in TRECVID’08

AU - Ren, Jinchang

AU - Jiang , J.

AU - Eckes, Christian

PY - 2008

Y1 - 2008

N2 - In this paper, our techniques used in TRECVID'08 on BBC rush summarization are described. Firstly, rush videos are hierarchical modeled using formal language description. Then, shot detection and V-unit determination are applied for video structuring; junk frames within the model are also effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to remove retakes. Then, each selected shot is ranked according to its length and sum of activity level for summarization. Competitive results have proved the effectiveness and efficiency of our techniques fully implemented in compressed-domain.

AB - In this paper, our techniques used in TRECVID'08 on BBC rush summarization are described. Firstly, rush videos are hierarchical modeled using formal language description. Then, shot detection and V-unit determination are applied for video structuring; junk frames within the model are also effectively removed. Thirdly, adaptive clustering is employed to group shots into clusters to remove retakes. Then, each selected shot is ranked according to its length and sum of activity level for summarization. Competitive results have proved the effectiveness and efficiency of our techniques fully implemented in compressed-domain.

KW - adaptive clustering

KW - hierarchical modelling

U2 - 10.1145/1463563.1463566

DO - 10.1145/1463563.1463566

M3 - Conference contribution book

SN - 978-1-60558-309-9

SP - 26

EP - 30

BT - TVS'08 Proc. of the 2nd ACM TRECVID Video Summarisation Workshop

T2 - TRECVid 2008

Y2 - 1 February 2008

ER -