Presentation 2007-07-10
Object Retrieval in Personal Eye-View Video using Adjacency
Takahiro KOIZUMI, Yuichi NAKAMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose an object retrieval technique from personal experience records. First, we present our object recognition method that detects object from shaky videos taken by a head mounted camera. It is, however, difficult to extract object areas and accurate features from the videos. We propose a retrieval method using objects relationship, which is associated by some kinds of contiguity. Even objects whose features are not accurately obtained can be retrieved through the relationships with other objects which features are easily obtained.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Personal Experience Record / Lifelog / Object Detection / Video Retrieval / Relevance Feedback / Contiguity / Adjacency
Paper # IE2007-29,MVE2007-32
Date of Issue

Conference Information
Committee IE
Conference Date 2007/7/2(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 Image Engineering (IE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Object Retrieval in Personal Eye-View Video using Adjacency
Sub Title (in English)
Keyword(1) Personal Experience Record
Keyword(2) Lifelog
Keyword(3) Object Detection
Keyword(4) Video Retrieval
Keyword(5) Relevance Feedback
Keyword(6) Contiguity
Keyword(7) Adjacency
1st Author's Name Takahiro KOIZUMI
1st Author's Affiliation Graduate School of Engineering, Kyoto University()
2nd Author's Name Yuichi NAKAMURA
2nd Author's Affiliation ACCMS, Kyoto University
Date 2007-07-10
Paper # IE2007-29,MVE2007-32
Volume (vol) vol.107
Number (no) 129
Page pp.pp.-
#Pages 6
Date of Issue