Presentation 2019-10-18
[Fellow Memorial Lecture] Knowledge Acquisition and Media Search Based on Crossmodal Information Processing
Kunio Kashino,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Media search can be roughly divided into two methods: a method in which fragments of media content are used as queries and corresponding parts are searched based on apparent matches and similarities of the content, and a method in which class labels and attribute information are used as queries and corresponding parts are searched based on matches and similarities of tags attached to media content. While each method is effectively utilized, so-called semantic gap has been a problem in the former in addition to speed and robustness of the search. And, in the latter, it has been a problem how to add tags to a large amount of content, and it is often difficult to add tags in advance comprehensively expecting what kind of queries will be made. To solve these problem, our research team has proposed the utilization of latent space mapping media content. In this talk, we review some of our research cases, especially focusing on crossmodal information processing, and discuss future possibilities with this approach.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) media search / crossmodal information processing / knowledge acquisition / semantic gap
Paper # PRMU2019-36
Date of Issue 2019-10-11 (PRMU)

Conference Information
Committee PRMU
Conference Date 2019/10/18(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Yoichi Sato(Univ. of Tokyo)
Vice Chair Toru Tamaki(Hiroshima Univ.) / Akisato Kimura(NTT)
Secretary Toru Tamaki(NTT) / Akisato Kimura(OMRON SINICX)
Assistant Yusuke Uchida(DeNA) / Takayoshi Yamashita(Chubu Univ.)

Paper Information
Registration To Technical Committee on Pattern Recognition and Media Understanding
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) [Fellow Memorial Lecture] Knowledge Acquisition and Media Search Based on Crossmodal Information Processing
Sub Title (in English)
Keyword(1) media search
Keyword(2) crossmodal information processing
Keyword(3) knowledge acquisition
Keyword(4) semantic gap
1st Author's Name Kunio Kashino
1st Author's Affiliation NTT(NTT)
Date 2019-10-18
Paper # PRMU2019-36
Volume (vol) vol.119
Number (no) PRMU-235
Page pp.pp.27-27(PRMU),
#Pages 1
Date of Issue 2019-10-11 (PRMU)