Presentation 2007-06-29
Fast Detection of Near-Duplicate Video Sequences by Division of Feature Space with Margin
Shugo SUZUKI, Tomokazu TAKAHASHI, Ichiro IDE, Hiroshi MURASE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recently, methods to search a video sequence from massively stored video data is being actively developed. As to broadcast video, a video sequence which is broadcasted repeatedly plays a special role, for example, a commercial or a video uesed in the same news topic. Detection of such duplicate video sequences, however, requires a large computational cost that is propotional to the square of whole video length. In this report, we propose a method that divides a feature space to detect duplicate video sequences rapidly. In the space division, dividing the space into two regions by a division boundary is executed hierarchically. In addition, to guarantee the same detection accuracy as that is obtained from the original space, a margin is set near the boundary. Since the experimental results showed that drastic reduction in computational time was achieved by the proposal method, the effectiveness in fast detection of duplicate video sequences was demonstrated.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Feature space division / duplicate video sequence / fast detection / accuracy guarantee
Paper # DE2007-10,PRMU2007-36
Date of Issue

Conference Information
Committee DE
Conference Date 2007/6/21(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Data Engineering (DE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Fast Detection of Near-Duplicate Video Sequences by Division of Feature Space with Margin
Sub Title (in English)
Keyword(1) Feature space division
Keyword(2) duplicate video sequence
Keyword(3) fast detection
Keyword(4) accuracy guarantee
1st Author's Name Shugo SUZUKI
1st Author's Affiliation Graduate School of Infomation Science, Nagoya University()
2nd Author's Name Tomokazu TAKAHASHI
2nd Author's Affiliation Graduate School of Infomation Science, Nagoya University
3rd Author's Name Ichiro IDE
3rd Author's Affiliation Graduate School of Infomation Science, Nagoya University:National Institute of Informatics
4th Author's Name Hiroshi MURASE
4th Author's Affiliation Graduate School of Infomation Science, Nagoya University
Date 2007-06-29
Paper # DE2007-10,PRMU2007-36
Volume (vol) vol.107
Number (no) 114
Page pp.pp.-
#Pages 5
Date of Issue