Presentation | 2007-12-14 Baseball Video Classification Using Training Code-Sequenced Scenes And Constraint Condition Based On Status of Game Takahiro MOCHIZUKI, Mahito FUJII, Nobuyuki YAGI, Kouichi SHINODA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper proposes a framework to apply the convential method for baseball scene classification into some event-classes (home-run, single-hit, walk, etc.) targeting at only digest videos by learning symbol-sequenced scenes based on the video patternization process to full-game videos, and a method to modify with information about game status in video intervals between events. "Event candidate scenes (ECS)" to be subjected the classification process are segmented based on apperance positions of pitch-view shot, replay scene, inning change and so on. Each ECS is classified into one event-class by the method for event classification based on learning patternized and symbol-sequenced scenes. Event scenes with no events (not-event scenes) are specified on the basis of relations between classification results for ECSs and their time length. We pick out information about game status from not-event scenes and removed intervals in ECS segmentation process by some ways including the detection of base region on the basis of image analysis, and event classfication results can be modified on some constraint conditions estabished by the game status. The proposed method is evaluated by an experiment using some Major League Baseball video games. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | baseball video / event classification / Hidden Markov Model / base detection |
Paper # | PRMU2007-149 |
Date of Issue |
Conference Information | |
Committee | PRMU |
---|---|
Conference Date | 2007/12/6(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 | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Baseball Video Classification Using Training Code-Sequenced Scenes And Constraint Condition Based On Status of Game |
Sub Title (in English) | |
Keyword(1) | baseball video |
Keyword(2) | event classification |
Keyword(3) | Hidden Markov Model |
Keyword(4) | base detection |
1st Author's Name | Takahiro MOCHIZUKI |
1st Author's Affiliation | NHK (Japan Broadcasting Corporation) Science and Technical Research Laboratories() |
2nd Author's Name | Mahito FUJII |
2nd Author's Affiliation | NHK (Japan Broadcasting Corporation) Science and Technical Research Laboratories |
3rd Author's Name | Nobuyuki YAGI |
3rd Author's Affiliation | NHK (Japan Broadcasting Corporation) Science and Technical Research Laboratories |
4th Author's Name | Kouichi SHINODA |
4th Author's Affiliation | Department of Computer Science, Tokyo Institute of Technology |
Date | 2007-12-14 |
Paper # | PRMU2007-149 |
Volume (vol) | vol.107 |
Number (no) | 384 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |