Presentation 2010-02-15
Detailed Estimation of Pitches in Baseball Games Based on Similar Image Retrieval on Motion Space-Time Images
Kyota AOKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) There are many videos about sports. In sports videos, the motions carry much information about shots. This paper proposes the video retrieval system based on the motions. We have many criteria about sameness. The method retrieves videos based on the similarity of the all motions described in frames. The method gets the motions using the reliable motion estimation method, and makes the 1-dimensional projections from the X motion, Y motion and the reliabilities. Resulting image has the 1-dimensional space axis and the 1-dimensional time axis. Using the images, the method retrieves shots as fast as the image retrievals do. This paper proposes the precise video retrieval method for baseball games based on the element motion retrieval and confirms the performance of the proposed method with the experiments about pitches that are most repeated plays in a baseball game.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) video retrieval / motion estimation / 1-dimensional projection
Paper # ITS2009-38,IE2009-132
Date of Issue

Conference Information
Committee ITS
Conference Date 2010/2/8(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 Intelligent Transport Systems Technology (ITS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Detailed Estimation of Pitches in Baseball Games Based on Similar Image Retrieval on Motion Space-Time Images
Sub Title (in English)
Keyword(1) video retrieval
Keyword(2) motion estimation
Keyword(3) 1-dimensional projection
1st Author's Name Kyota AOKI
1st Author's Affiliation School of Engineering, Utsunomiya University()
Date 2010-02-15
Paper # ITS2009-38,IE2009-132
Volume (vol) vol.109
Number (no) 414
Page pp.pp.-
#Pages 6
Date of Issue