| 電子情報通信学会 研究会発表申込システム 講演論文 詳細 |
技報閲覧サービス [ログイン] 技報アーカイブ |
| トップに戻る | 前のページに戻る | [Japanese] / [English] |
| 講演抄録／キーワード | ||
| 講演名 | 2024-01-25 16:10 ［ポスター講演］Automated Electronic Strike Zone and Pitch Trajectory Overlay System Based on AI computer vision ○Lin Wang・Bor-Yao Tseng・Hung-Tse Chiang・Jiann-Liang Chen（NTUST）・Han-Chuan Hsieh（TISS）  IA2023-64 |
|
| 抄録 | （和） | At present, the electronic strike zone in practical applications is usually fixed in a specific height range, and the accuracy of the electronic strike zone can be improved through human pose estimation. Therefore, we proposed a system, using the common baseball game perspective image and AI visual recognition technology, automatically detects the height and posture of the player's body, and combined with the baseball nine-grid positioning principle, timely establish the strike zone in line with the baseball rules and physical characteristics of the player. Regarding Intersection over Union (IoU) metrics, our system achieved an accuracy of 0.8541, which is 14.29% higher than the 0.7473 accuracy of the current electronic strike zone. Considering the spectator experience and assistance in pitcher training, our system has also added the function of automatic detection of pitch tracks, which can overlay multiple pitch tracks and visually show the difference between pitch tracks. Combined with our proposed strike zone range detection system, the overlapping relationship between pitch trajectory and strike zone can be demonstrated more comprehensively. |
| （英） | At present, the electronic strike zone in practical applications is usually fixed in a specific height range, and the accuracy of the electronic strike zone can be improved through human pose estimation. Therefore, we proposed a system, using the common baseball game perspective image and AI visual recognition technology, automatically detects the height and posture of the player's body, and combined with the baseball nine-grid positioning principle, timely establish the strike zone in line with the baseball rules and physical characteristics of the player. Regarding Intersection over Union (IoU) metrics, our system achieved an accuracy of 0.8541, which is 14.29% higher than the 0.7473 accuracy of the current electronic strike zone. Considering the spectator experience and assistance in pitcher training, our system has also added the function of automatic detection of pitch tracks, which can overlay multiple pitch tracks and visually show the difference between pitch tracks. Combined with our proposed strike zone range detection system, the overlapping relationship between pitch trajectory and strike zone can be demonstrated more comprehensively. | |
| キーワード | （和） | Pose estimation / Electronic Strike Zone / Ball Trajectory / AI Computer Vision / / / / |
| （英） | Pose estimation / Electronic Strike Zone / Ball Trajectory / AI Computer Vision / / / / | |
| 文献情報 | 信学技報, vol. 123, no. 372, IA2023-64, pp. 40-44, 2024年1月. | |
| 資料番号 | IA2023-64 | |
| 発行日 | 2024-01-18 (IA) | |
| ISSN | Online edition: ISSN 2432-6380 | |
| 著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します．(許諾番号：10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) | |
| PDFダウンロード | IA2023-64 |
|
| 研究会情報 | |
| 研究会 | IA |
| 開催期間 | 2024-01-25 - 2024-01-25 |
| 開催地（和） | 関西学院大学 丸の内キャンパス |
| 開催地（英） | Kwansei Gakuin Univiversity, Marunouchi Campus |
| テーマ（和） | センサー ネットワーク, IoT, M2M, 一般, 及び IA2023 - Workshop on Internet Architecture and Applications 2023 |
| テーマ（英） | Sensor Network, IoT, M2M, etc., and IA2023 - Workshop on Internet Architecture and Applications 2023 |
| 講演論文情報の詳細 | |
| 申込み研究会 | IA |
| 会議コード | 2024-01-IA |
| 本文の言語 | 英語 |
| タイトル（和） | |
| サブタイトル（和） | |
| タイトル（英） | Automated Electronic Strike Zone and Pitch Trajectory Overlay System Based on AI computer vision |
| サブタイトル（英） | |
| キーワード(1)（和/英） | Pose estimation / Pose estimation |
| キーワード(2)（和/英） | Electronic Strike Zone / Electronic Strike Zone |
| キーワード(3)（和/英） | Ball Trajectory / Ball Trajectory |
| キーワード(4)（和/英） | AI Computer Vision / AI Computer Vision |
| キーワード(5)（和/英） | / |
| キーワード(6)（和/英） | / |
| キーワード(7)（和/英） | / |
| キーワード(8)（和/英） | / |
| 第1著者 氏名（和/英/ヨミ） | Lin Wang / Lin Wang / |
| 第1著者 所属（和/英） | National Taiwan University of Science and Technology (略称： NTUST) National Taiwan University of Science and Technology (略称： NTUST) |
| 第2著者 氏名（和/英/ヨミ） | Bor-Yao Tseng / Bor-Yao Tseng / |
| 第2著者 所属（和/英） | National Taiwan University of Science and Technology (略称： NTUST) National Taiwan University of Science and Technology (略称： NTUST) |
| 第3著者 氏名（和/英/ヨミ） | Hung-Tse Chiang / Hung-Tse Chiang / |
| 第3著者 所属（和/英） | National Taiwan University of Science and Technology (略称： NTUST) National Taiwan University of Science and Technology (略称： NTUST) |
| 第4著者 氏名（和/英/ヨミ） | Jiann-Liang Chen / Jiann-Liang Chen / |
| 第4著者 所属（和/英） | National Taiwan University of Science and Technology (略称： NTUST) National Taiwan University of Science and Technology (略称： NTUST) |
| 第5著者 氏名（和/英/ヨミ） | Han-Chuan Hsieh / Han-Chuan Hsieh / |
| 第5著者 所属（和/英） | Taiwan Institute of Sports Science (略称： TISS) Taiwan Institute of Sports Science (略称： TISS) |
| 講演者 | 第1著者 |
| 発表日時 | 2024-01-25 16:10:00 |
| 発表時間 | 30分 |
| 申込先研究会 | IA |
| 資料番号 | IA2023-64 |
| 巻番号（vol） | vol.123 |
| 号番号（no） | no.372 |
| ページ範囲 | pp.40-44 |
| ページ数 | 5 |
| 発行日 | 2024-01-18 (IA) |
[研究会発表申込システムのトップページに戻る]