Presentation 2007-10-25
Video Classification Using Linear Subspace Methods(International Session 1)
Kei KIKUCHI, Seiji HOTTA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a video classifier using linear subspace methods. In the proposed method, the subspace of a training video is formed with color histograms of each frame. When an input video is given, the total sum of projection distances between all frames of the input video and the linear subspace of a training video is computed. This total sum of distances can be regarded as the distance between the input video and the training one. Thus, the input video is classified into the class to which the nearest subspace belongs. By the proposed method, we can classify videos without clustering or shot segmentation. In addition, the learning algorithms of the proposed video classification are presented for reducing memory costs. Furthermore, we propose a shot searching method using subspaces for detecting desired shots or harmful contents. The performance of the proposed method is verified with experiments on short videos downloaded from the Web.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) video classification / linear subspace method / learning subspace method / shot searching
Paper # PRMU2007-93
Date of Issue

Conference Information
Committee PRMU
Conference Date 2007/10/18(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 Pattern Recognition and Media Understanding (PRMU)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Video Classification Using Linear Subspace Methods(International Session 1)
Sub Title (in English)
Keyword(1) video classification
Keyword(2) linear subspace method
Keyword(3) learning subspace method
Keyword(4) shot searching
1st Author's Name Kei KIKUCHI
1st Author's Affiliation Tokyo University of Agriculture and Technology()
2nd Author's Name Seiji HOTTA
2nd Author's Affiliation Tokyo University of Agriculture and Technology
Date 2007-10-25
Paper # PRMU2007-93
Volume (vol) vol.107
Number (no) 281
Page pp.pp.-
#Pages 6
Date of Issue