高橋 翔 (タカハシ シヨウ)

工学研究院 土木工学部門 先端社会システム准教授
数理・データサイエンス教育研究センター准教授
Last Updated :2024/12/06

■研究者基本情報

学位

  • 博士(情報科学), 北海道大学, 2013年03月

プロフィール情報

  • 【学歴】
    平成18年3月 木更津工業高等専門学校情報工学科 卒業.
    平成20年3月 北海道大学工学部情報工学科 卒業.
    平成22年3月 北海道大学大学院情報科学研究科メディアネットワーク専攻 修士課程 修了.
    平成25年3月 北海道大学大学院情報科学研究科メディアネットワーク専攻 博士課程 修了.

    【職歴】
    平成22年4月 北海学園大学 工学部 非常勤講師.
    平成23年4月 日本学術振興会 特別研究員.
    平成25年4月 北海道大学大学院情報科学研究科 特任助教.
    平成29年7月 北海道大学数理・データサイエンス教育研究センター 特任准教授.
    平成30年4月 北海道大学大学院工学研究院 准教授.
    平成30年9月-平成31年3月 フィレンツェ大学 客員教授.

    【所属学会】
    IEEE,電子情報通信学会,映像情報メディア学会,土木学会,自動車技術会,交通工学研究会,日本環境共生学会 各会員.博士(情報科学).

Researchmap個人ページ

研究キーワード

  • 道路・交通データ集積
  • 支援システム
  • スポーツ映像
  • 可視化
  • 意味解析
  • 映像処理

研究分野

  • 社会基盤(土木・建築・防災), 土木計画学、交通工学
  • 情報通信, 学習支援システム
  • 情報通信, ヒューマンインタフェース、インタラクション
  • 情報通信, データベース

■経歴

経歴

  • 2018年04月 - 現在
    北海道大学, 大学院工学研究院, 准教授
  • 2018年09月 - 2019年03月
    フィレンツェ大学, Media Integration and Communication Center, 客員教授
  • 2017年07月 - 2018年03月
    北海道大学, 数理・データサイエンス教育研究センター, 特任准教授
  • 2013年04月 - 2017年06月
    北海道大学, 大学院情報科学研究科, 特任助教
  • 2011年04月 - 2013年03月
    日本学術振興会, 特別研究員
  • 2010年04月 - 2012年03月
    北海学園大学, 工学部, 非常勤講師

学歴

  • 2010年04月 - 2013年03月, 北海道大学, 大学院情報科学研究科, メディアネットワーク専攻博士後期課程, 日本国
  • 2008年04月 - 2010年03月, 北海道大学, 大学院情報科学研究科, メディアネットワーク専攻修士課程, 日本国
  • 2006年04月 - 2008年03月, 北海道大学, 工学部, 情報工学科, 日本国
  • 2001年04月 - 2006年03月, 木更津工業高等専門学校, 情報工学科, 日本国

委員歴

  • 2023年10月 - 現在
    北海道開発局 札幌開発建設部 一般国道5号創成川通 防災計画・施設検討会, 委員, 政府
  • 2023年06月 - 現在
    北海道開発局 札幌開発建設部 国道12号白石本通第二電線共同溝PFI事業 有識者等委員会, 委員, 政府
  • 2022年09月 - 現在
    北海道開発局 札幌開発建設部 創成トンネル浸水対策等技術検討会, 委員, 政府
  • 2020年09月 - 現在
    地域道路経済戦略研究会 北海道地方研究会, 委員, 政府
  • 2020年06月 - 現在
    北海道大規模小売店舗立地審議会, 特別委員, 自治体
  • 2020年04月 - 現在
    北海道開発局 稚内開発建設部 総合審査評価委員会, 委員, 政府
  • 2019年06月 - 現在
    北海道土木技術会道路研究会, 委員兼幹事, その他
  • 2020年11月 - 2023年03月
    北海道積雪寒冷対応システム検討会, 座長, 自治体
  • 2022年04月 - 2023年02月
    交通工学研究会 第4回JSTEシンポジウム運営小委員会, 幹事, 学協会
  • 2022年07月 - 2022年10月
    IEEE Global Conference on Consumer Electronics (GCCE 2022), Technical Program Committee Chair, 学協会
  • 2021年01月 - 2021年12月
    IEEE Global Conference on Consumer Electronics (GCCE 2021), Conference Chair, 学協会
  • 2019年04月 - 2021年03月
    土木学会北海道支部, 論文担当幹事・チーフ幹事, 学協会
  • 2020年04月 - 2021年02月
    交通工学研究会 第2回JSTEシンポジウム運営小委員会, 幹事, 学協会
  • 2020年01月 - 2021年01月
    International Conference on Pattern Recognition (ICPR 2020), Associate Editor, 学協会
  • 2019年01月 - 2019年12月
    日本環境共生学会 第22回(2019年度)学術大会, 実行委員, 学協会

■研究活動情報

受賞

  • 2023年01月, International Workshop on Advanced Image Technology 2023, Best Paper Award               
    Improvement of Nighttime Visibility Estimation Based on > Spatio-Temporal Correlation in Videos
    Masahiro Yagi;Sho Takahashi;Toru Hagiwara
  • 2022年12月, 土木学会, AI・データサイエンス論文賞               
    複数の車載センサーデータを統合した冬期の路面状態のLate Fusion > による推定モデル
    石附将武;高橋翔;萩原亨;石井啓太;岩﨑悠志;森徹平;花塚泰史
  • 2022年10月, 交通工学研究会, 第42回交通工学研究発表会 研究奨励賞               
    歩行者の安心かつ円滑な横断を目的とした自動運転車による意思伝達装置に関する研究
    和田駿一;高橋翔;萩原亨
  • 2021年, 映像情報メディア学会, 丹羽高柳賞論文賞               
    Multimodal Important Scene Detection in Far-view Soccer Videos Based on Single Deep Neural Architecture
    Tomoki Haruyama;Sho Takahashi;Takahiro Ogawa;Miki Haseyama
  • 2020年10月, 2020 IEEE 9th Global Conference on Consumer Electronics, IEEE GCCE 2020 Excellent Poster Award               
    An Estimation Method of Road Narrowing Condition in in-Vehicle Camera
    Kozo Okumura;Sho Takahashi;Toru Hagiwara
  • 2020年09月, IEEE International Conference on Consumer Electronics-Taiwan, Best Paper Award               
    An Estimation Method of Visibility Level on Winter Road Based on Multiple Features in CCTV Images
    Shotaro Kawata;Sho Takahashi;Toru Hagiwara
  • 2019年10月, IEEE 8th Global Conference on Consumer Electronics (GCCE), IEEE GCCE 2019 Excellent Poster Award               
    Masahiro Yagi;Sho Takahashi;Toru Hagiwara
  • 2019年10月, IEEE 8th Global Conference on Consumer Electronics (GCCE), IEEE GCCE 2019 Excellent Student Paper Award               
    Takayuki Abe;Sho Takahashi;Toru Hagiwara
  • 2019年10月, IEEE 8th Global Conference on Consumer Electronics (GCCE), IEEE GCCE 2019 Excellent Demo! Award, Gold Prize               
    Sho Takahashi;Masahiro Yagi;Toru Hagiwara
  • 2018年10月, IEEE 7th Global Conference on Consumer Electronics (GCCE), 1st Prize IEEE GCCE 2018 Excellent Poster Award               
    Tomoki Haruyama;Sho Takahashi;Takahiro Ogawa;Miki Haseyama
  • 2017年12月, 映像情報メディア学会, 優秀研究発表賞               
    高橋 翔
  • 2013年10月, IEEE 2nd Global Conference on Consumer Electronics, IEEE GCCE 2013 Outstanding Poster Award               
    高橋 翔
  • 2009年11月, 平成21年度電気・情報関係学会北海道支部連合大会, 優秀論文発表賞               
    高橋 翔

論文

  • AR-based Merging Assistance at Expressway and Its Verification
    Sho Takahashi, Ryohei Maruyama, Toru Hagiwara
    International Journal of Intelligent Transportation Systems Research, 2024年12月
    研究論文(学術雑誌)
  • Drivers' collision avoidance behavior: Timing of pedestrian detection when turning right at signalized intersections
    Chinami Fukui, Sho Takahashi, Toru Hagiwara
    Asian Transport Studies, 2024年
    研究論文(学術雑誌)
  • Macroscopic Fundamental Diagram理論に基づく冬期の降雪強度が札幌市中心部の交通流に与える影響に関する研究
    山城皓太郎, 萩原亨, 高橋翔
    交通工学研究発表会論文集(Web), 43rd, 2023年
  • 複数の車載センサーデータを統合した冬期の路面状態のLate Fusionによる推定モデル               
    石附将武, 高橋翔, 萩原亨, 石井啓太, 岩﨑悠志, 森徹平, 花塚泰史
    AI・データサイエンス論文集, 2022年11月, [査読有り], [責任著者]
    日本語, 研究論文(学術雑誌)
  • 予測誤差補正によるLSTMを用いた歩行軌跡予測の高精度化に関する研究               
    鴨藤功武, 高橋翔, 萩原亨
    AI・データサイエンス論文集, 2022年11月, [査読有り], [責任著者]
    日本語, 研究論文(学術雑誌)
  • 複数識別器の確信度に基づくLate-fusionによる車載カメラ映像を用いた夜間の視界レベル推定               
    佐藤諒, 高橋翔, 萩原亨, 永田泰浩, 大橋一仁
    AI・データサイエンス論文集, 2022年11月, [査読有り], [責任著者]
    日本語, 研究論文(学術雑誌)
  • Development of Road Visibility Inspection System Using Driving Video Images Recorded by On-board Video Camera               
    Kazuhito OHASHI, Yasuhiro NAGATA, Yasuhiro KANEDA, Toru HAGIWARA, Sho TAKAHASHI, Yuki NAKAMURA
    ROUTES/ROADS Magazine, RR393-028, 2022年06月, [査読有り]
    英語, 研究論文(学術雑誌)
  • 空間周波数および深層学習に基づく機械学習による視程レベル推定 -特徴選択手法による空間周波数の選択効果-               
    高橋翔, 河田祥太朗, 萩原亨
    土木学会論文誌 D3(土木計画学), 77, 5, I_1077, I_1084, 2022年05月, [査読有り], [筆頭著者]
    日本語, 研究論文(学術雑誌)
  • Development of Poor Visibility Assessment Method in Winter Using Images Taken by On-Board Video Camera               
    Yuki Nakamura, Toru Hagiwara, Yasuhiro Nagata, Sho Takahashi
    Journal of EASTS, 18, 1824, 1841, 2022年03月, [査読有り], [最終著者]
    英語, 研究論文(学術雑誌)
  • Effects of Time-Gap Settings of Adaptive Cruise Control (ACC) on Driver’s Risk Feeling Estimated by the Car-Following Situation               
    Shuhei Wada, Sho Takahashi, Tomonori Ohiro, Kazunori Munehiro, Minoru Okada, Toru Hagiwara
    Journal of EASTS, 14, 2133, 2148, 2022年03月, [査読有り]
    英語, 研究論文(学術雑誌)
  • Awareness of the prevention through design (PtD) concept among design engineers in the Philippines
    Rimmon Labadan, Kriengsak Panuwatwanich, Sho Takahashi
    Engineering Management in Production and Services, 14, 1, 78, 92, Walter de Gruyter GmbH, 2022年03月01日, [査読有り]
    英語, 研究論文(学術雑誌), Abstract

    The “Prevention through Design” (PtD) concept considers construction safety during the design process. Several countries are currently practising PtD, including the UK, Singapore, Malaysia, Australia, and the USA, which is still not the case in the Philippines. The study presented in this paper aimed to indicate the current level of awareness of the PtD concept among the structural engineers and purposed to generate a basis of initiatives to introduce or improve the understanding and adoption of PtD in the Philippines. A knowledge, attitude, and practice (KAP) questionnaire was distributed to survey respondents selected through a snowball sampling method, consisting of structural engineers currently working in the Philippines. Sixty-one (61) structural engineers responded and were analysed in this study. Results indicated that PtD was relatively a new concept for most structural engineers in the Philippines. Similarly, the designers’ knowledge of the concept was still low. However, structural engineers viewed PtD as necessary and its implementation as essential in the construction industry. Despite the known concerns in the PtD implementation, structural engineers favoured the adoption of the concept. The paper also discussed challenges and key drivers for implementing PtD in the Philippines based on the questionnaire results and supporting literature reviews. The findings and methodology presented in this paper could serve as a baseline for a larger sample size covering other design trades, such as architectural, electrical, and mechanical design services leading to the broader adoption of PtD in the Philippines. Furthermore, the framework of this study could also apply to other countries with similar contexts.
  • 高速道路合流部の交通円滑化を支援する速度誘導灯に関する研究               
    大石侑亮, 河合レナ, 高橋翔, 萩原亨
    交通工学論文集, 8, 2, A_54, A_61, 2022年02月, [査読有り]
    日本語, 研究論文(学術雑誌)
  • Virtual Reality Driving Simulator を用いた市街地交差点における右折ドライバの横断歩行者認知に関する研究               
    岡崎泰勢, 高橋翔, 丸山凌平, 萩原亨
    交通工学論文集, 8, 2, A_185, A_193, 2022年02月, [査読有り]
    日本語, 研究論文(学術雑誌)
  • Image and CAIS Features-Based Estimation of Road Surface Condition on Winter Local Road.
    Masamu Ishizuki, Sho Takahashi, Toru Hagiwara, Keita Ishii, Yuji Iwasaki, Teppei Mori, Yasushi Hanatsuka
    11th IEEE Global Conference on Consumer Electronics(GCCE), 79, 80, IEEE, 2022年
    研究論文(国際会議プロシーディングス)
  • A Virtual Reality Driving Simulator with Gaze Tracking for Analyzing Driver's Behavior.
    Ryohei Maruyama, Sho Takahashi, Toru Hagiwara
    LifeTech, 144, 145, 2022年, [査読有り]
    英語, 研究論文(国際会議プロシーディングス)
  • An Estimation Method of Visibility Level Based on Multiple Models Using In-vehicle Camera Videos under Nighttime.
    Ryo Sato, Masahiro Yagi, Sho Takahashi, Toru Hagiwara, Yasuhiro Nagata, Kazuhito Ohashi
    GCCE, 530, 531, 2021年, [査読有り]
    研究論文(国際会議プロシーディングス)
  • Vehicle Behavior Measurement based-on RTK-GNSS for Driver's Risk Feeling Estimation.
    Sho Takahashi, Shuhei Wada, Toru Hagiwara, Kazunori Munehiro
    GCCE, 528, 529, 2021年, [査読有り], [筆頭著者]
    研究論文(国際会議プロシーディングス)
  • LSTM-Based Prediction Method of Crowd Behavior for Robust to Pedestrian Detection Error.
    Isamu Kamoto, Takayuki Abe, Sho Takahashi, Toru Hagiwara
    GCCE, 218, 219, 2021年, [査読有り]
    研究論文(国際会議プロシーディングス)
  • An Estimation Method of Road Surface Condition on Winter Expressway via Mobile Nets using In-vehicle Camera Images.
    Tomoyuki Takase, Sho Takahashi, Toru Hagiwara, Tomonori Ohiro, Yuji Iwasaki, Teppei Mori, Yasushi Hanatsuka
    GCCE, 109, 110, 2021年, [査読有り]
    研究論文(国際会議プロシーディングス)
  • An Edge Computing System for Estimating Road Surface Condition on Winter Expressway.
    Masahiro Yagi, Tomoyuki Takase, Sho Takahashi, Toru Hagiwara, Tomonori Ohiro, Yuji Iwasaki, Teppei Mori, Yasushi Hanatsuka
    GCCE, 71, 72, 2021年, [査読有り]
    研究論文(国際会議プロシーディングス)
  • Advanced Road Visibility Inspection System for Winter Road Maintenance Using Microcomputer.
    Sho Takahashi, Toru Hagiwara, Ryo Sato, Kazuhito Ohashi, Yasuhiro Nagata
    GCCE, 28, 29, 2021年, [査読有り]
    研究論文(国際会議プロシーディングス)
  • 冬期道路環境へのACC(Adaptive Cruise Control)の適応条件に関する基礎的研究
    和田脩平, 高橋翔, 白石直之, 宗広一徳, 岡田稔, 内藤利幸, 萩原亨
    交通工学論文集(特集号), 7, 2, A_289, A_297, 一般社団法人 交通工学研究会, 2021年, [査読有り]
    日本語,

    ACCで冬期道路を走行しているドライバの先行車に対する主観的リスク認知レベルを実車で計測し、冬期道路におけるACCの適応条件の検証を試みた。ドライバがACCを用いて先行車を追従中、先行車が減速し接近する場面を想定した。その結果、ACC走行時のドライバの主観的リスク認知レベルは通常運転に比べて大きくなった。また、先行車との接近時におけるTHW(Time Headway)とTTC(Time To Collision)を説明変数、主観的なリスク認知レベルを目的変数とする統計的モデルを推定した。モデルはTHWのみが有意となった。モデルと主観的なリスク認知レベルの結果から、ドライバが圧雪路面をACCで走行するときにTHW設定を長くすることで、通常運転と同等のリスク認知レベルとなる可能性が示唆された。

  • ACC(Adaptive Cruise Control)利用時における速度誘導灯を用いた高速道路合流部でのドライバの錯綜回避行動に関する研究
    河合レナ, 萩原亨, 高橋翔, 寺倉嘉宏, 大石 侑亮
    交通工学論文集(特集号), 7, 2, A_316, A_325, 一般社団法人 交通工学研究会, 2021年, [査読有り]
    日本語, 研究論文(学術雑誌),

    本研究では,高速道路本線で ACC を用いた運転を行っているドライバの合流車との錯綜を軽減するために,事前の速度調整装置として走行車線中央に埋め込んだ速度誘導灯を提案した.ドライビングシミュレータを用いて 46 名の実験参加者で走行実験を行い,走行記録・主観評価などから速度誘導灯の意図に関する説明の違いがドライバの運転行動に影響を与えることが示された.その中で,速度誘導灯の意図を説明した実験参加者は,合流部のノーズ端の上流で速度誘導灯の指示に合わせた適切な速度調整を選択していた.これによりスムーズな合流となり,ドライバが感じる合流車による主観的な危険度も低くなった.加えて,速度誘導灯を点灯させていない時における本線車のドライバの加減速行動から,速度誘導灯の設置が効果的となる位置を明らかにした.

  • 高速道路合流部におけるARを用いたHMIによる合流行動支援とその有効性の検証
    丸山凌平, 高橋翔, 萩原亨, 寺倉嘉宏
    ヒューマンインタフェース学会論文誌, 23, 1, 19, 28, ヒューマンインタフェース学会, 2021年, [査読有り]
    日本語, 研究論文(学術雑誌), This study tries to support merging behavior by providing information about the behavior of the mainline vehicle as a next-generation Human-Machine Interface (HMI) based on Augmented Reality (AR). Also, in this paper, we verify the effectiveness of visual information on the driver using AR by analyzing the gaze. Specifically, this paper provides information on the timing in which the merging vehicle cannot merge by using AR to the driver. This information is generated based on the position and speed of the mainline vehicle. In this study, a Virtual Reality (VR) driving simulator that can measure the gaze was developed for experiments. By utilizing the VR driving simulator, driving experiments that provide various kinds of information including AR to the merging vehicle was performed. Also, by using the obtained data at the driving experiments, the driver's gazes are analyzed. In this gaze analysis, we verify the burden on the driver of checking the mainline is reduced by the information provision. Furthermore, the timing of the merging is measured to evaluate the safety and smoothness of the merging. The experimental results suggested that providing the unsuitable timing of merging by AR enables the driver to understand easily the condition of the mainline.
  • ACCによるカーブ走行時における冬期路面の影響に関する研究
    白石直之, 高橋翔, 萩原亨, 岡田稔, 内藤利幸, 宗広一徳
    土木学会論文誌 D3(土木計画学), 76, 5, I_1409, I_1416, 公益社団法人 土木学会, 2021年, [査読有り]
    日本語,

    冬期のSAEレベル2以上の運転支援・自動運転導入にはドライバが危険を感じる走行環境の把握が必要である.本研究では公道において走行調査を行い,カーブ区間に着目し冬期道路環境でACC動作中にドライバがとった危険回避行動(調査参加者:6名)と,そのときの道路線形,路面のすべりやすさ,気象条件について検討した.走行調査の結果から道路線形と路面のすべり抵抗値,走行速度を組み合わせた条件を考え,道路線形が厳しい環境であってもすべり抵抗値が高い環境であればドライバの危険感が減少し,ドライバによる危険回避行動を減らすことができる可能性を明らかにした.また,道路線形が厳しい区間やすべり抵抗値が低い区間ではシステムが予め走行速度を低下させることでドライバに危険感を与えない走行が可能となることを示唆した.

  • AVOIDANCE BEHAVIOR ON BICYCLE TRIPS DETECTION BASED ON MULTIMODAL APPROACH FOR ROAD MANAGEMENT
    Masahiro Yagi, Sho Takahashi, Toru Hagiwara
    Journal of JSCE D3(Infrastructure Planning and Management), 76, 5, I_899, I_907, 2021年, [査読有り]
    研究論文(学術雑誌)
  • 自転車の回避行動に関するデータ集積のためのエッジコンピューティングシステムの構築
    八木 雅大, 高橋 翔, 萩原 亨
    土木学会論文集D3 (土木計画学), 76, 5, I_859, I_867, 公益社団法人 土木学会, 2021年, [査読有り]
    日本語,

    現在,CCTV や車載カメラにより取得された多くの道路映像が,道路管理者の元に集積されている.しかしながら,その多くは取得された画像や映像がそのまま集積されており,ネットワークを介した伝送や保存のための容量が膨大となっている.また,変状や障害物が存在する道路状況では,自転車は通常の走行が困難となり,それらを回避する走行を強いられる.したがって,そのような道路状況にあることを把握可能とすることは,より適切な道路管理につながる.そこで,本稿では,自転車の回避行動に関するデータをエッジコンピューティングによって,データ量を削減しながら集積するシステムを提案する.本稿の最後では,実験によって,搭載するアルゴリズムの有効性およびシステムの有用性について確認する.

  • User-selectable event summarization in unedited raw soccer video via multimodal bidirectional LSTM
    Tomoki Haruyama, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    ITE Transactions on Media Technology and Applications, 9, 1, 42, 53, 2021年, [査読有り]
    研究論文(学術雑誌), A new method that generates user-selectable event summaries from unedited raw soccer videos is presented in this paper. Since there are more unedited raw soccer videos than broadcasted/distributed soccer videos and unedited videos have various viewers, it is necessary to analyze these videos for meeting the demands of various viewers. The proposed method introduces a multimodal CNN-BiLSTM architecture for analyzing unedited raw soccer videos. This architecture extracts candidate scenes for event summarization from unedited soccer videos and classifies these scenes into typical events. Finally, our method generates user-selectable event summaries by simultaneously considering the importance of candidate scenes and the event classification results. Experimental results using real unedited raw soccer videos show the effectiveness of our method.
  • Deterioration level estimation via neural network maximizing category-based ordinally supervised multi-view canonical correlation.
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    Multim. Tools Appl., 80, 15, 23091, 23112, 2021年, [査読有り]
    研究論文(学術雑誌), A deterioration level estimation method via neural network maximizing category-based ordinally supervised multi-view canonical correlation is presented in this paper. This paper focuses on real world data such as industrial applications and has two contributions. First, a novel neural network handling multi-modal features transforms original features into features effectively representing deterioration levels in transmission towers, which are one of the infrastructures, with consideration of only correlation maximization. It can be realized by setting projection matrices maximizing correlations between multiple features into weights of hidden layers. That is, since the proposed network has only a few hidden layers, it can be trained from a small amount of training data. Second, since there exist diverse characteristics and an ordinal scale in deterioration levels, the proposed method newly derives category-based ordinally supervised multi-view canonical correlation analysis (Co-sMVCCA). Co-sMVCCA enables estimation of effective projection considering both within-class divergence and the ordinal scale between classes. Experimental results showed that the proposed method realizes accurate deterioration level estimation.
  • A Study on Avoidance Behavior on Bicycle Trips Detection Using Multiple Features for Improvement Road Management
    Masahiro Yagi, Sho Takahashi, Toru Hagiwara
    2020 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020, 2020年09月28日, [査読有り]
    研究論文(国際会議プロシーディングス), This paper proposes a method for detecting avoidance behavior on bicycle trips in videos using multiple features. Our previous method detected the avoidance behavior on bicycle trips by using single feature based on rider's body parts. Also, the collaborative use of multiple features is effective for video analysis tasks. Thus, the proposed method introduces two new features for detection performance improvement. The introduced features are based on optical flow and Convolutional Neural Network (CNN), which are widely used in video analysis. Consequently, more accurate detection is realized. Experimental results show the effectiveness of the proposed method.
  • An Estimation Method of Candidate Region for Superimposing Information Based on Gaze Tracking Data in Soccer Videos
    Genki Suzuki, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    2020 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020, 2020年09月28日, [査読有り]
    研究論文(国際会議プロシーディングス), A novel method estimating candidate regions for superimposing information in soccer videos based on gaze tracking data is presented in this paper. The proposed method generates a likelihood map based on visual attention regions based on the gaze tracking data and detection results of objects such as players and soccer goals in soccer videos. Candidate regions for superimposing information are estimated by using the likelihood map. Experimental results show that the proposed method realizes effective candidate region estimation.
  • Automatic Generation of Training Data for Deep Learning in AR-based Transportation Support System
    Takayuki Abe, Sho Takahashi, Toru Hagiwara
    2020 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020, 2020年09月28日, [査読有り]
    研究論文(国際会議プロシーディングス), In this paper, we propose a method that automatically generates training data necessary for construction of deep learning-based model which supports indication of traffic data on augmented reality (AR). In the case of the data indication to road user by AR, there is a risk that important objects are hidden by the data. Thus, locations of the data indication should be adaptively determined. Therefore, accurate determination of the locations which recognizes the important objects via deep learning is expected. However, for deep learning, a large amounts of training data need to be prepared. In our method, the large amount of training data is automatically generated. Therefore, supporting location determination of the data indication via deep learning is realized. Experimental results show the effectiveness of our method.
  • An Estimation Method of Visibility Level on Winter Road Based on Multiple Features in CCTV Images
    Shotaro Kawata, Sho Takahashi, Toru Hagiwara
    2020 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020, 2020年09月28日, [査読有り]
    研究論文(国際会議プロシーディングス), Since drivers acquire a lot of information with eyes, the estimation of the level of visibility (visibility level) on the winter roads contributes toward enhancing traffic safety. Therefore, this paper proposes a method for estimating the visibility level on the winter roads from closed-circuit television (CCTV) images. In the proposed method, the visibility level of each image is estimated based on features acquired via Fourier transform and Convolutional Neural Network (CNN). Specifically, by constructing support vector machines (SVMs) for each feature, the probabilities of the visibility level are calculated. Furthermore, based on a comparison of the calculated probabilities of SVMs, the proposed method estimates the visibility level accurately. Experimental results show the effectiveness of the proposed method.
  • An Estimation Method of Visibility Level Based on Low Rank Matrix Completion Using GPV Data.
    Gen Ohkama, Sho Takahashi, Toru Hagiwara
    IEEE International Conference on Consumer Electronics - Taiwan(ICCE-TW), 1, 2, IEEE, 2020年
    研究論文(国際会議プロシーディングス)
  • 視線データと点検データの正準相関に基づく道路橋点検のための類似点検データ検索
    前田圭介, 斉藤僚汰, 髙橋翔, 小川貴弘, 長谷山美紀
    土木学会論文集F3(土木情報学), 76, 1, 74, 76, 2020年, [査読有り]
  • Similar scene retrieval in soccer videos with weak annotations by multimodal use of bidirectional LSTM.
    Tomoki Haruyama, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    The 2nd ACM International Conference on Multimedia in Asia, 27, 8, ACM, 2020年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper presents a novel method to retrieve similar scenes in soccer videos with weak annotations via multimodal use of bidirectional long short-term memory (BiLSTM). The significant increase in the number of different types of soccer videos with the development of technology brings valid assets for effective coaching, but it also increases the work of players and training staff. We tackle this problem with a nontraditional combination of pre-trained models for feature extraction and BiLSTMs for feature transformation. By using the pre-trained models, no training data is required for feature extraction. Then effective feature transformation for similarity calculation is performed by applying BiLSTM trained with weak annotations. This transformation allows for highly accurate capture of soccer video context from less annotation work. In this paper, we achieve an accurate retrieval of similar scenes by multimodal use of this BiLSTM-based transformer trainable with less human effort. The effectiveness of our method was verified by comparative experiments with state-of-the-art using actual soccer video dataset.
  • Feature Integration Via Geometrical Supervised Multi-View Multi-Label Canonical Correlation For Incomplete Label Assignment.
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    Proceedings - International Conference on Image Processing, ICIP, 2020-October, 46, 50, IEEE, 2020年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper presents feature integration via geometrical supervised multi-view multi-label canonical correlation analysis (GSM2CCA) for incomplete label assignment. The problem of incomplete labels is frequently encountered in the multi-label classification problem where the training labels are obtained via crowd-sourcing. In such a situation, consideration of only the label correlation, which is the basic approach, is not suitable for improvement of representation ability of features. For dealing with the incomplete label assignment, GSM2CCA constructs effective feature embedding space providing the discriminant ability by introducing both the multi-label correlation and feature similarity of the original feature space into its objective function. Since novel integrated features with high discriminant ability can be calculated by our GSM2CCA, performance improvement of multi-label classification with the incomplete label assignment is realized. The main contribution of this paper is the realization of the effective feature integration via the adoption of the combination use of label similarity and locality preserving projection of heterogeneous features for solving the problem of the incomplete label assignment. The effectiveness of GSM2CCA by applying GSM2CCA-based feature integration to heterogeneous features calculated from various convolutional neural network models is verified via experimental results.
  • An Estimation Method of Road Narrowing Condition in In-vehicle Camera.
    Kozo Okumura, Sho Takahashi, Toru Hagiwara
    2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020, 733, 734, IEEE, 2020年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper proposes an estimation method of road narrowing condition by piled snow in in-vehicle camera image. The winter roads are possibly narrowed with the piled snow and caused traffic congestion. To maintain the urban functions the data accumulation of piled snow is very useful for snow removal planning. Therefore, in this paper, a novel method for estimating the road narrowing condition from in in-vehicle camera image is proposed. The proposed method is composed by hierarchical two steps which identify roads which have enough road capacity with piled snow and narrow roads which cause traffic congestion. These steps are used the low level image features and the position of road structures. In order to verify the effectiveness of the proposed method, we conduct two experiments. Experimental results show the effectiveness of our method.
  • Visibility Level Estimation in Winter CCTV Images Based on Decision Level Fusion Using Logistic Regression.
    Shotaro Kawata, Sho Takahashi, Toru Hagiwara
    2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020, 640, 641, IEEE, 2020年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper proposes a method for estimating the degree of visibility (visibility level) in winter closed-circuit television (CCTV) images by decision level fusion based on logistic regression (LR). The proposed method classifies on the basis of two support vector machines (SVMs) classifiers and fuses the classification results by utilizing LRbased late fusion. In the proposed method, the SVMs which tentatively estimate visibility are constructed based on each CCTV image feature that represents the contrast of images and neuron values of the neural network. Also, the proposed method evaluates the visibility based on LR model trained by using SVM outputs. The effectiveness of our method is verified from experiments by utilizing actual CCTV images.
  • An Estimation Method of Visibility Level Based on Low Rank Matrix Completion Using Positional Relationship of GPV Data.
    Gen Ohkama, Sho Takahashi, Toru Hagiwara
    2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020, 638, 639, IEEE, 2020年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper proposes a method for estimating visibility level based on low rank matrix completion using the positional relationship of GPV data. On a snowy day, the driving tasks can be hard because visibility is poor. By providing visibility level to drivers and road manager, safer traffic will be realized. To estimate visibility, weather data, which are significantly related to visibility, are used in the proposed method. In the proposed method, Grid Point Value (GPV) data and visibility data including missing values on unobserved points are generated to matrix corresponding to the actual location. The missing values are estimated based on low rank matrix completion. The effectiveness of our method are shown by utilizing actual data.
  • [Papers] Multimodal Important Scene Detection in Far-view Soccer Videos Based on Single Deep Neural Architecture
    Haruyama Tomoki, Takahashi Sho, Ogawa Takahiro, Haseyama Miki
    映像情報メディア学会英語論文誌, 8, 2, 89, 99, 一般社団法人 映像情報メディア学会, 2020年, [査読有り]
    英語, 研究論文(学術雑誌),

    The details of the matches of soccer can be estimated from visual and audio sequences, and they correspond to the occurrence of important scenes. Therefore, the use of these sequences is suitable for important scene detection. In this paper, a new multimodal method for important scene detection from visual and audio sequences in far-view soccer videos based on a single deep neural architecture is presented. A unique point of our method is that multiple classifiers can be realized by a single deep neural architecture that includes a Convolutional Neural Network-based feature extractor and a Support Vector Machine-based classifier. This approach provides a solution to the problem of not being able to simultaneously optimize different multiple deep neural architectures from a small amount of training data. Then we monitor confidence measures output from this architecture for the multimodal data and enable their integration to obtain the final classification result.

  • [Paper] A Method for Player Importance Prediction from Player Network Using Gaze Position Estimated by LSTM
    Suzuki Genki, Takahashi Sho, Ogawa Takahiro, Haseyama Miki
    映像情報メディア学会英語論文誌, 8, 3, 151, 160, 一般社団法人 映像情報メディア学会, 2020年, [査読有り]
    英語, 研究論文(学術雑誌),

    A novel method for player importance prediction from a player network using gaze positions estimated by Long Short-Term Memory (LSTM) in soccer videos is presented in this paper. By newly using an estimation model of gaze positions trained by gaze tracking data of experienced persons, it is expected that the importance of each player can be predicted. First, we generate a player network by utilizing the estimated gaze positions and first-arrival regions representing players' connections, e.g., passes between players. The gaze positions are estimated by LSTM that is newly trained from the gaze tracking data of experienced persons. Second, the proposed method predicts the importance of each player by applying the Hypertext Induced Topic Selection (HITS) algorithm to the constructed network. Consequently, prediction of the importance of each player based on soccer tactic knowledge of experienced persons can be realized without constantly obtaining gaze tracking data.

  • Estimation of Interest Levels From Behavior Features via Tensor Completion Including Adaptive Similar User Selection
    Keisuke Maeda, Tetsuya Kushima, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    IEEE Access, 8, 126109, 126118, Institute of Electrical and Electronics Engineers ({IEEE}), 2020年, [査読有り]
    研究論文(学術雑誌), A method for estimating interest levels from behavior features via tensor completion including adaptive similar user selection is presented in this paper. The proposed method focuses on a tensor that is suitable for data containing multiple contexts and constructs a third-order tensor in which three modes are 'products', 'users' and 'user behaviors and interest levels' for these products. By complementing this tensor, unknown interest level estimation of a product for a target user becomes feasible. For further improving the estimation performance, the proposed method adaptively selects similar users for the target user by focusing on converged estimation errors between estimated interest levels and known interest levels in the tensor completion. Furthermore, the proposed method can adaptively estimate the unknown interest from the similar users. This is the main contribution of this paper. Therefore, the influence of users having different interests is reduced, and accurate interest level estimation can be realized. In order to verify the effectiveness of the proposed method, we show experimental results obtained by estimating interest levels of users holding books.
  • Retrieval of similar scenes based on multimodal distance metric learning in soccer videos
    Tomoki Haruyama, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    MMSports 2019 - Proceedings of the 2nd International Workshop on Multimedia Content Analysis in Sports, co-located with MM 2019, 10, 15, ACM, 2019年10月15日, [査読有り]
    研究論文(国際会議プロシーディングス), © 2019 Association for Computing Machinery. This paper presents a new method for retrieval of similar scenes based on multimodal distance metric learning in far-view soccer videos that broadly capture soccer fields and are not edited. We extract visual features and audio features from soccer video clips, and we extract text features from text data corresponding to these soccer video clips. In addition, distance metric learning based on Laplacian Regularized Metric Learning is performed to calculate the distances for each kind of features. Finally, by determining the final rank by integrating these distances, we realize successful multimodal retrieval of similar scenes from query scenes of soccer video clips. Experimental results show the effectiveness of our retrieval method.
  • Neural Network Maximizing Ordinally Supervised Multi-View Canonical Correlation for Deterioration Level Estimation
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    Proceedings - International Conference on Image Processing, ICIP, 2019-September, 919, 923, IEEE, 2019年09月, [査読有り]
    研究論文(国際会議プロシーディングス), © 2019 IEEE. This paper presents a neural network maximizing ordinally supervised multi-view canonical correlation for deterioration level estimation. The contributions of this paper are twofold. First, in order to calculate features representing deterioration levels on transmission towers, which is one of the infrastructures, a novel neural network handling multi-modal features is constructed from a small amount of training data. Specifically, in our method, effective transformation to features with high discriminant ability without using many hidden layers is realized by setting projection matrices maximizing correlation between multiple features into hidden layer's weights. Second, since there exists ordinal scale in deterioration levels, the proposed method newly derives ordinally supervised multi-view canonical correlation analysis (OsMVCCA). OsMVCCA enables estimation of the effective projection considering not only label information but also their ordinal scales. Experimental results show that the proposed method realizes accurate deterioration level estimation.
  • 認識・抽出 センサから得られる視聴行動データを活用したユーザの関心推定の高度化
    長谷山 美紀, 小川 貴弘, 髙橋 翔, 原川 良介
    画像ラボ, 30, 7, 8, 12, 日本工業出版, 2019年07月
    日本語
  • Estimation of users' interest levels using tensor completion with SemiCCA
    Tetsuya Kushima, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    2019 IEEE 1st Global Conference on Life Sciences and Technologies, LifeTech 2019, 239, 240, IEEE, 2019年03月, [査読有り]
    研究論文(国際会議プロシーディングス), © 2019 IEEE. This paper presents a new method for estimation of users' interest levels using tensor completion with SemiCCA. The proposed method extracts new features maximizing correlation between features calculated from partially paired users' behavior and contents with semi-supervised canonical correlation analysis (SemiCCA). By this approach, we can successfully use the contents that users have not viewed for the interest level estimation. Moreover, our method utilizes the tensor completion to estimate unknown interest levels. Consequently, in the proposed method, accurate estimation of interest levels using SemiCCA and the tensor completion is realized. Experimental results are shown to verify the effectiveness of the proposed method by using actual data.
  • Multi-feature Fusion Based on Supervised Multi-view Multi-label Canonical Correlation Projection.
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2019, Brighton, United Kingdom, May 12-17, 2019, 2019-May, 3936, 3940, IEEE, 2019年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper presents multi-feature fusion based on supervised multi-view multi-label canonical correlation projection (sM2CP). The proposed method applies sM2CP-based feature fusion to multiple features obtained from various convolutional neural networks (CNNs) whose characteristics are different. Since new fused features with high representation ability can be obtained, performance improvement of multi-label classification is realized. Specifically, in order to tackle the multi-label problem, sM2CP introduces a label similarity information of label vectors into the objective function of supervised multi-view canonical correlation analysis. Thus, sM2CP can deal with complex label information such as multi-label annotation. The main contribution of this paper is the realization of feature fusion of multiple CNN features for the multi-label problem by introducing multi-label similarity information into the canonical correlation analysis-based feature fusion approach. Experimental results show the effectiveness of sM2CP, which enables effective fusion of multiple CNN features.
  • Convolutional sparse coding-based deep random vector functional link network for distress classification of road structures.
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    Comput. Aided Civ. Infrastructure Eng., 34, 8, 654, 676, 2019年, [査読有り]
    英語, 研究論文(学術雑誌), This paper presents a convolutional sparse coding (CSC)-based deep random vector functional link network (CSDRN) for distress classification of road structures. The main contribution of this paper is the introduction of CSC into a feature extraction scheme in the distress classification. CSC can extract visual features representing characteristics of target images because it can successfully estimate optimal convolutional dictionary filters and sparse features as visual features by training from a small number of distress images. The optimal dictionaries trained from distress images have basic components of visual characteristics such as edge and line information of distress images. Furthermore, sparse feature maps estimated on the basis of the dictionaries represent both strength of the basic components and location information of regions having their components, and these maps can represent distress images. That is, sparse feature maps can extract key components from distress images that have diverse visual characteristics. Therefore, CSC-based feature extraction is effective for training from a limited number of distress images that have diverse visual characteristics. The construction of a novel neural network, CSDRN, by the use of a combination of CSC-based feature extraction and the DRN classifier, which can also be trained from a small dataset, is shown in this paper. Accurate distress classification is realized via the CSDRN.
  • Multimodal Retrieval of Similar Soccer Videos Based on Optimal Combination of Multiple Distance Measures.
    Tomoki Haruyama, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019, 665, 666, IEEE, 2019年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper presents a new multimodal method for retrieval of similar soccer videos based on optimal combination of multiple distance measures. Our method first extracts three types of Convolutional Neural Network-based features focusing the players' actions, the audience's cheers and prompt reports. Then, by applying the optimal distance measure to each feature, we calculate the similarities between a query video and videos in a database. Finally, we realize accurate retrieval of similar soccer videos by integrating these similarities. Experiments on actual soccer videos demonstrate encouraging results.
  • A Calculation Method of Degree of Data Indication Regions in First-Person View Videos for Improvement Transportation.
    Takayuki Abe, Sho Takahashi, Toru Hagiwara
    2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019, 558, 559, IEEE, 2019年, [査読有り]
    研究論文(国際会議プロシーディングス), In this paper, we propose a calculation method of degree for detecting regions of various data indication in first-person view videos. In recent years, contents of augmented reality (AR) have been implemented due to advances in computer hardware and interface. By indicating the data on road user's view, intelligent transport systems (ITS) is enhanced. However, since the user's view includes a lot of visually important information which the user should perceive, for indicating the data on the user's view, it is necessary to detect regions that do not include the information. In our method, degree as criteria is obtained based on seam carving and saliency map model. By utilizing the degree, it is possible to adaptively detect the regions of the data indication from different viewpoints. Experimental results show the effectiveness of our method.
  • Similarity Calculation Based on Pass Regions in Soccer Videos.
    Sho Takahashi, Marco Bertini, Alberto Del Bimbo, Miki Haseyama, Toru Hagiwara
    2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019, 515, 516, IEEE, 2019年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper discusses a similarity calculation between soccer scenes in soccer videos. In the soccer games, pass regions, which are parts of the soccer court that include some pass courses are very important elements of each tactic in various scenes. Therefore, by visualizing the pass regions on soccer court in soccer videos, many audiences can easily understand tactics and game situations. Also, since the visualized pass regions describe tactics of soccer games, tactically similar scenes include similar pass regions each other. Thus, the similarity of soccer scenes is obtained by utilizing estimated pass regions in soccer videos. In this paper, we discuss the similarity calculation for soccer scenes. Specifically, the similarity of soccer scenes is compared which are calculated from a Deep Neural Network(DNN)-based visual feature of visualized pass regions, a DNN-based visual feature of original soccer image and Active Net-based feature of estimated pass regions.
  • An Evaluation Method of Obstacle Avoidance Behavior on Bicycle Trip Using Rider's Gesture.
    Masahiro Yagi, Sho Takahashi, Toru Hagiwara
    2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019, 513, 514, IEEE, 2019年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper proposes a method for evaluation of obstacle avoidance behavior on bicycle trips in videos using rider's gesture based on OpenPose. Since a bicycle is a vehicle which is ridden by rotating a wheel with load transfer of a rider, there is a high correlation between bicycle behavior and rider's gesture. Therefore, we evaluate the obstacle avoidance behavior based on rider's gesture. In the proposed method, features about rider's gesture are obtained by utilizing OpenPose. In addition, support vector machines (SVMs) are constructed based on the features. Our method detects the obstacle avoidance behavior and classifies that to a degree of danger. Experimental results show the effectiveness of our method.
  • Analysis of Various Web Data for Visualization of Travel Time and Accessibility.
    Kozo Okumura, Sho Takahashi, Toru Hagiwara
    2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019, 439, 440, IEEE, 2019年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper proposes a method for the calculation of travel time and the visualization of accessibility based on multiple data sources. In an environment where transport data are not fully open, construction of novel transportation systems which are planned based on enough data is limited. Our method calculates travel time based on data of multiple sources, which are not fully corrected data and visualizes novel accessibility by defining a novel index based on the travel time. The index is calculated from variance and mean of the travel time. As a result, a visual comparison of the accessibility between each area is realized by utilizing the proposed method. Thus, integrating some transport data is expected to contribute to constructing the optimal transportation system. Experimental results show the effectiveness of our evaluation method.
  • Data Accumulation System of Obstacle Avoidance Behavior on Bicycle Trip for Transportation Engineering.
    Sho Takahashi, Masahiro Yagi, Toru Hagiwara
    2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019, 311, 312, IEEE, 2019年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper proposes a data accumulation system of obstacle avoidance behavior on bicycle trips in videos using gesture based on deep learning-based features. In the field of transportation engineering, realization of image/video processing-based methods for analyzing of various situations in transportation scenes has been expected. Therefore, many videos taken by CCTV and dashcams are accumulated in various transportation organizations. However, since these accumulates only raw videos which are huge size, that transmission, accumulate device and analysis of accumulated videos are very high cost as social infrastructures. Therefore, this paper constructs a system for accumulating small-sized videos, extracted features and analyzed results of the transportation scene, which is the obstacle avoidance behavior on bicycle trips. This system is constructed on a small size computer which can be worked by using the battery in order to equip on various structures. Also, an algorithm for detecting obstacle avoidance behavior on bicycle trips in videos are realized by utilizing deep learning-based features and support vector machines.
  • Interest Level Estimation Based on Tensor Completion via Feature Integration for Partially Paired User’s Behavior and Videos
    Kushima Tetsuya, Takahashi Sho, Ogawa Takahiro, Haseyama Miki
    IEEE Access, 7, 148576, 148585, 2019年, [査読有り]
    英語, 研究論文(学術雑誌), A novel method for interest level estimation based on tensor completion via feature integration for partially paired users' behavior and videos is presented in this paper. The proposed method defines a novel canonical correlation analysis (CCA) framework that is suitable for interest level estimation, which is a hybrid version of semi-supervised CCA (SemiCCA) and supervised locality preserving CCA (SLPCCA) called semi-supervised locality preserving CCA (S2LPCCA). For partially paired users' behavior and videos in actual shops and on the Internet, new integrated features that maximize the correlation between partially paired samples by the principal component analysis (PCA)-mixed CCA framework are calculated. Then videos that users have not watched can be used for the estimation of users' interest levels. Furthermore, local structures of partially paired samples in the same class are preserved for accurate estimation of interest levels. Tensor completion, which can be applied to three contexts, videos, users and 'canonical features and interest levels,' is used for estimation of interest levels. Consequently, the proposed method realizes accurate estimation of users' interest levels based on S2LPCCA and the tensor completion from partially paired training features of users' behavior and videos. Experimental results obtained by applying the proposed method to actual data show the effectiveness of the proposed method.
  • Team Tactics Estimation in Soccer Videos Based on a Deep Extreme Learning Machine and Characteristics of the Tactics.
    Genki Suzuki, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    IEEE Access, 7, 153238, 153248, 2019年, [査読有り]
    英語, 研究論文(学術雑誌), A novel method for estimating team tactics in soccer videos based on a Deep Extreme Learning Machine (DELM) and unique characteristics of tactics is presented in this paper. The proposed method estimates the tactics of each team from players' formations and enables successful training from a limited amount of training data. Specifically, the estimation of tactics consists of two stages. First, by utilizing two DELMs corresponding to the two teams, the proposed method estimates the provisional tactics of each team. Second, the proposed method updates the team tactics based on unique characteristics of soccer tactics, the relationship between tactics of the two teams and information on ball possession. Consequently, since the proposed method estimates the team tactics that satisfy these characteristics, accurate estimation results can be obtained. In an experiment, the proposed method is applied to actual soccer videos to verify its effectiveness.
  • Estimation of Deterioration Levels of Transmission Towers via Deep Learning Maximizing Canonical Correlation Between Heterogeneous Features
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    IEEE Journal of Selected Topics in Signal Processing, 12, 4, 633, 644, Institute of Electrical and Electronics Engineers (IEEE), 2018年08月, [査読有り]
    英語, 研究論文(学術雑誌), This paper presents estimation of deterioration levels of transmission towers via deep learning maximizing the canonical correlation between heterogeneous features. In the proposed method, we newly construct a correlation-maximizing deep extreme learning machine (CMDELM) based on a local receptive field (LRF). For accurate deterioration level estimation, it is necessary to obtain semantic information that effectively represents deterioration levels. However, since the amount of training data for transmission towers is small, it is difficult to perform feature transformation by using many hidden layers such as general deep learning methods. In CMDELM-LRF, one hidden layer, which maximizes the canonical correlation between visual features and text features obtained from inspection text data, is newly inserted. Specifically, by using projections obtained by maximizing the canonical correlation as weight parameters of the hidden layer, feature transformation for extracting semantic information is realized without designing many hidden layers. This is the main contribution of this paper. Consequently, CMDELM-LRF realizes accurate deterioration level estimation from a small amount of training data.
  • Distress classification of class-imbalanced inspection data via correlation-maximizing weighted extreme learning machine
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    Advanced Engineering Informatics, 37, 79, 87, Elsevier Ltd, 2018年08月01日, [査読有り]
    英語, 研究論文(学術雑誌), This paper presents distress classification of class-imbalanced inspection data via correlation-maximizing weighted extreme learning machine (CMWELM). For distress classification, it is necessary to extract semantic features that can effectively distinguish multiple kinds of distress from a small amount of class-imbalanced data. In recent machine learning techniques such as general deep learning methods, since effective feature transformation from visual features to semantic features can be realized by using multiple hidden layers, a large amount of training data are required. However, since the amount of training data of civil structures becomes small, it becomes difficult to perform successful transformation by using these multiple hidden layers. On the other hand, CMWELM consists of two hidden layers. The first hidden layer performs feature transformation, which can directly extract the semantic features from visual features, and the second hidden layer performs classification with solving the class-imbalanced problem. Specifically, in the first hidden layer, the feature transformation is realized by using projections obtained by maximizing the canonical correlation between visual and text features as weight parameters of the hidden layer without designing multiple hidden layers. Furthermore, the second hidden layer enables successful training of our classifier by using weighting factors concerning the class-imbalanced problem. Consequently, CMWELM realizes accurate distress classification from a small amount of class-imbalanced data.
  • Automatic estimation of deterioration level on transmission towers via deep extreme learning machine based on local receptive field
    Maeda Keisuke, Takahashi Sho, Ogawa Takahiro, Haseyama Miki
    IEEE Conference Proceedings, 2017-, ICIP, 2379, 2383, IEEE Computer Society, 2018年02月20日, [査読有り]
    英語, 研究論文(国際会議プロシーディングス), This paper presents an automatic estimation method of deterioration levels on transmission towers via Deep Extreme Learning Machine based on Local Receptive Field (DELM-LRF). Although Convolutional Neural Network (CNN) requires a large number of training images, it is difficult to prepare a sufficient number of training images of transmission towers. Thus, we generate a novel estimation method which enables training from a small number of training images. Specifically, we automatically extract image features based on Local Receptive Field (LRF) which combines convolution and pooling without using hand-craft features and estimate deterioration levels via Deep Extreme Learning Machine (DELM), which is a part of efficient deep learning methods. The derivation of DELM-LRF is the biggest contribution of this paper, and it can be trained from less training images compared to CNN. Experimental results show the effectiveness of DELM-LRF for the estimation of deterioration levels on transmission towers. Consequently, the proposed method makes it possible to approach challenging tasks with high expertise having difficulty in preparing enough images.
  • Estimation of Important Scenes in Soccer Videos Based on Collaborative Use of Audio-Visual CNN Features.
    Tomoki Haruyama, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    IEEE 7th Global Conference on Consumer Electronics, GCCE 2018, Nara, Japan, October 9-12, 2018, 710, 711, IEEE, 2018年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper presents a novel method for estimating important scenes in soccer videos based on collaborative use of audio-visual Convolutional Neural Network (CNN) features. In soccer games, since game situations influence not only players' movements but also audiences' cheers, analyses of their audio and visual sequences are useful for the estimation of important scenes. In our method, such scenes are estimated from audio and visual CNN features via support vector machine (SVM) in each feature. Furthermore, by applying weighted majority voting based on confidences defined from the SVM-based estimation results, accurate estimation of important scenes becomes feasible. Experimental results show the effectiveness of our method.
  • Team Tactics Estimation in Soccer Videos via Deep Extreme Learning Machine Based on Players Formation.
    Genki Suzuki, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    IEEE 7th Global Conference on Consumer Electronics, GCCE 2018, Nara, Japan, October 9-12, 2018, 116, 117, IEEE, 2018年, [査読有り]
    研究論文(国際会議プロシーディングス), A method of team tactics estimation in soccer videos is presented in this paper. Our method enables estimation of basic tactics in each team on the basis of the Deep-Extreme Learning Machine (DELM) by using features of players formation. In the soccer games, team tactics relate to each other team. Therefore, the proposed method obtains final estimation results by utilizing two DELMs of each team and their relationship. Since the proposed method takes into consideration the relevance of the estimated tactics in each team, we realize accurate tactics estibimation. Experimental results using actual soccer videos showed the effectiveness of our method.
  • Binary Sparse Representation Based on Arbitrary Quality Metrics and Its Applications
    OGAWA Takahiro, TAKAHASHI Sho, WADA Naofumi, TANAKA Akira, HASEYAMA Miki
    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 101, 11, 1776, 1785, 電子情報通信学会, 2018年, [査読有り]
    英語, 研究論文(学術雑誌), <p>Binary sparse representation based on arbitrary quality metrics and its applications are presented in this paper. The novelties of the proposed method are twofold. First, the proposed method newly derives sparse representation for which representation coefficients are binary values, and this enables selection of arbitrary image quality metrics. This new sparse representation can generate quality metric-independent subspaces with simplification of the calculation procedures. Second, visual saliency is used in the proposed method for pooling the quality values obtained for all of the parts within target images. This approach enables visually pleasant approximation of the target images more successfully. By introducing the above two novel approaches, successful image approximation considering human perception becomes feasible. Since the proposed method can provide lower-dimensional subspaces that are obtained by better image quality metrics, realization of several image reconstruction tasks can be expected. Experimental results showed high performance of the proposed method in terms of two image reconstruction tasks, image inpainting and super-resolution.</p>
  • Interest Level Estimation of Items via Matrix Completion Based on Adaptive User Matrix Construction.
    Tetsuya Kushima, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    2018 IEEE International Conference on Multimedia and Expo, ICME 2018, San Diego, CA, USA, July 23-27, 2018, 2018-July, 1, 6, IEEE Computer Society, 2018年, [査読有り]
    研究論文(国際会議プロシーディングス), This paper presents a novel method for interest level estimation of items via matrix completion based on adaptive user matrix construction. The proposed method introduces a new criterion for adaptively constructing a user matrix that consists of user behavior features and interest levels, which are evaluated by target users and similar users. In the estimation, the matrix completion via rank minimization using the truncated nuclear norm is applied to the constructed matrix. The proposed method enables both of the interest level estimation of the target users and the selection of the similar users suitable for the estimation by monitoring errors caused in the matrix completion algorithm. The caused errors indicate the minimum differences between the estimated interest levels and true ones, and they can be regarded as the criterion for both of the optimal estimation and the adaptive selection. Furthermore, the proposed method uses weight matrices for decreasing an influence of missing data on the estimation. Consequently, accurate estimation of the interest levels becomes feasible by using the adaptively constructed matrix. Experimental results obtained by applying the proposed method to users' behavior and interest data show the effectiveness of the proposed method.
  • 熟練技術者の判定に基づいた道路橋における類似点検データの検索
    斉藤 僚汰, 高橋 翔, 小川 貴弘, 長谷山 美紀
    土木学会論文集F3(土木情報学), 74, 1, 67, 77, 土木学会, 2018年, [査読有り]
    日本語, 研究論文(学術雑誌), This paper presents a retrieval method of similar inspection records in road structures based on metric learning using experienced inspectors' evaluation. Inspection records of road structures include images and text-based information such as category of distress, damaged parts and degree of damage. The proposed method calculates distances from query inspection records, and rank lists of retrieval results are obtained for each feature. In this approach, the distance quantification are updated on the basis of experienced inspectors' evaluation. Finally, the proposed method obtains retrieval results by integrating the multiple rank lists. The experimental results show the effectiveness of the proposed method.
  • Interest level estimation based on matrix completion via rank minimization
    Tetsuya Kushima, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    2017 IEEE 6th Global Conference on Consumer Electronics, GCCE 2017, 2017-, GCCE, 1, 2, Institute of Electrical and Electronics Engineers Inc., 2017年12月19日, [査読有り]
    英語, 研究論文(国際会議プロシーディングス), This paper presents a novel method for interest level estimation based on matrix completion via rank minimization. The proposed method estimates interest levels of target objects from human behavior features which are extracted during selecting these objects. Specifically, by adopting matrix completion via rank minimization, unknown interest levels can be estimated. Furthermore, the proposed method can also estimate unknown interest levels with some missing behavior features which are not correctly extracted by sensors. Experimental results show the effectiveness of the proposed method.
  • Distress Classification of Road Structures via Adaptive Bayesian Network Model Selection
    K. Maeda, S. Takahashi, T. Ogawa, M. Haseyama
    JOURNAL OF COMPUTING IN CIVIL ENGINEERING, 31, 5, ASCE-AMER SOC CIVIL ENGINEERS, 2017年09月, [査読有り]
    英語, 研究論文(学術雑誌), This paper presents an accurate distress classification method via adaptive Bayesian network model selection for maintenance inspection of road structures. The main contribution of this paper is adaptive selection of two Bayesian network models concerning classification performance. The proposed method trains a tag-based Bayesian network model based on inspection items and estimates its classification performance. Furthermore, for distresses that degrade the classification performance of the tag-based Bayesian network model, the proposed method trains another multifeature Bayesian network model based on inspection items and distress images. Consequently, the proposed method can adaptively select optimal Bayesian network models according to the estimated performance of the tag-based Bayesian network model. In actual maintenance inspection, distresses are generally classified either from inspection items alone or from both inspection items and visual information of distress images-i.e., distress classification has two patterns. Therefore the adaptive model selection approach is suitable for this classification scheme. Experimental results show that the proposed method outperforms several comparative methods and is suitable for actual maintenance inspection due to its low computation costs. (C) 2017 American Society of Civil Engineers.
  • Biomimetics Image Retrieval Platform
    Miki Haseyama, Takahiro Ogawa, Sho Takahashi, Shuhei Nomura, Masatsugu Shimomura
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, E100D, 8, 1563, 1573, IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG, 2017年08月, [査読有り], [招待有り]
    英語, 研究論文(学術雑誌), Biomimetics is a new research field that creates innovation through the collaboration of different existing research fields. However, the collaboration, i.e., the exchange of deep knowledge between different research fields, is difficult for several reasons such as differences in technical terms used in different fields. In order to overcome this problem, we have developed a new retrieval platform, "Biomimetics image retrieval platform," using a visualization-based image retrieval technique. A biological database contains a large volume of image data, and by taking advantage of these image data, we are able to overcome limitations of text-only information retrieval. By realizing such a retrieval platform that does not depend on technical terms, individual biological databases of various species can be integrated. This will allow not only the use of data for the study of various species by researchers in different biological fields but also access for a wide range of researchers in fields ranging from materials science, mechanical engineering and manufacturing. Therefore, our platform provides a new path bridging different fields and will contribute to the development of biomimetics since it can overcome the limitation of the traditional retrieval platform.
  • Deterioration Level Estimation on Transmission Towers via Extreme Learning Machine based on Combination Use of Local Receptive Field and Principal Component Analysis               
    K. Maeda, S. Takahashi, T. Ogawa, M. Haseyama
    International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC), 457, 458, 2017年07月, [査読有り]
    英語, 研究論文(国際会議プロシーディングス)
  • Distress Classification of Class Imbalanced Data for Maintenance Inspection of Road Structures in Express Way               
    K. Maeda, S. Takahashi, T. Ogawa, M. Haseyama
    International Conference on Civil and Building Engineering Informatics in conjunction with Conference on Computer Applications in Civil and Hydraulic Engineering (ICCBEI & CCACHE), 182, 185, 2017年04月, [査読有り]
    英語, 研究論文(国際会議プロシーディングス)
  • Distress classification of road structures via decision level fusion
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    International Conference on Digital Signal Processing, DSP, 2016, DSP, 589, 593, Institute of Electrical and Electronics Engineers Inc., 2017年03月01日, [査読有り]
    英語, 研究論文(国際会議プロシーディングス), A distress classification method of road structures via decision level fusion is presented in this paper. In order to classify various kinds of distresses accurately, the proposed method integrates multiple classification results with considering their performance, and this is the biggest contribution of this paper. By introducing this approach, it becomes feasible to adaptively integrate the multiple classification results based on the accuracy of each classifier for a target sample. Consequently, realization of the accurate distress classification can be expected. Experimental results show that our method outperforms existing methods.
  • A method of important player extraction based on link analysis in soccer videos
    Sho Takahashi, Miki Haseyama
    ITE Transactions on Media Technology and Applications, 5, 2, 42, 48, Institute of Image Information and Television Engineers, 2017年, [査読有り]
    英語, 研究論文(学術雑誌), In this paper, a method for extraction of important players in soccer videos based on link analysis is proposed. In a soccer match, players perform shoot tackles, assistance, and covering. Furthermore, the soccer tactics are defined the formation of players based on various relationships between players. The proposed method extracts the important players, in order to obtain information for understanding the soccer matches for various audiences. Specifically, our method notes that relationship between players, who cooperate with each other by the pass and the covering, is similar to relationship between web pages which are connected by links. First, the proposed method obtains player networks based on relationship between players in each team. The relationships are defined based on player positions and the possibility of the pass or the covering between players. Finally, in the proposed method, by applying the link analysis to the obtained player networks, important players are extracted. By realizing this approach, important players are extracted from the player networks based on the possibility of the pass or the covering between players. In the last of this paper, the above link analysis-based method was applied to actual soccer matches to show the reasonability of our method.
  • サッカー映像における試合内容の理解を促すデータの可視化               
    高橋 翔, 長谷山 美紀
    映像情報メディア学会誌, 70, 5, 722, 724, 2016年09月, [査読有り]
    日本語
  • 堤防点検における技術者の視線データと熟練度の分析に関する一考察 (画像工学)
    高橋 翔, 三改木 裕矢, 小川 貴弘
    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 115, 459, 177, 180, 電子情報通信学会, 2016年02月22日
    日本語
  • Decision Level Fusion-based Team Tactics Estimation in Soccer Videos
    Genki Suzuki, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    2016 IEEE 5TH GLOBAL CONFERENCE ON CONSUMER ELECTRONICS, 1, 2, IEEE, 2016年, [査読有り]
    英語, 研究論文(国際会議プロシーディングス), A decision-level fusion (DLF)-based team tactics estimation method in soccer videos is newly presented. In our method, tactics estimation based on audio-visual and formation features is newly adopted since the tactics of the soccer game are closely related to the audio-visual sequences and player positions. Therefore, by using these features, we classify the tactics via Support. Vector Machine (SVM). Furthermore, by applying DIA' to the SVM-based classification results, the two modalities are integrated to obtain more accurate tactics estimation results. Some results of experiments verify the superiority of our method.
  • Distress Classification of Road Structures via Multiple Classifier-based Bayesian Network               
    K. Maeda, S. Takahashi, T. Ogawa, M. Haseyama
    International Workshop on Advanced Image Technology (IWAIT), 1, 4, 2016年, [査読有り]
    英語, 研究論文(国際会議プロシーディングス)
  • DLF-based speech segment detection and its application to audio noise removal for video conferences
    Kazuto Sasaki, Takahiro Ogawa, Sho Takahashi, Miki Haseyama
    ITE Transactions on Media Technology and Applications, 4, 1, 68, 77, Institute of Image Information and Television Engineers, 2016年, [査読有り]
    英語, 研究論文(学術雑誌), A new decision-level fusion (DLF)-based speech segment detection method and its application to audio noise removal for video conferences are presented in this paper. The proposed method calculates visual and audio features from video sequences and audio signals, respectively, obtained in video conferences. Features extracted from mouth regions of participants and attribution degrees of speech class are used as visual and audio features, respectively, and Support Vector Machine (SVM)-based classification is performed by using each kind of feature. The SVM classifier performs two-class classification of speech and non-speech segments to realize speech segment detection. From the detection results obtained from the visual and audio features, DLF based on Supervised Learning from Multiple Experts is performed to successfully obtain the final detection results with focus on the accuracy of each detection result. Then, from audio signals in the non-speech segments detected by our method, we can extract noise information to realize accurate audio noise removal in the speech segments.
  • 6. サッカー映像における試合内容の理解を促すデータの可視化
    高橋 翔, 長谷山 美紀
    映像情報メディア学会誌, 70, 9, 722, 724, 一般社団法人 映像情報メディア学会, 2016年
    日本語, 研究論文(学術雑誌)
  • Improvement of video coding efficiency based on sparse contractive mapping approach
    Zaixing He, Takahiro Ogawa, Sho Takahashi, Miki Haseyama, Xinyue Zhao
    NEUROCOMPUTING, 173, 1898, 1907, ELSEVIER SCIENCE BV, 2016年01月, [査読有り]
    英語, 研究論文(学術雑誌), This paper presents a new method for improving video coding efficiency based on a sparse contractive mapping approach. The proposed method introduces a new sparse contractive mapping approach to replace the traditional intra prediction in the video coding standards such as H.264/AVC. Specifically, the intra- and its following inter-frame are respectively approximated by the sparse representation, satisfying contractive mapping. Then these two frames can be reconstructed from an arbitraryinitial image by utilizing a few representation coefficients. With this advantage, the proposed method reduces the total amount of bits by removing MBs in the target I frame, whose approximation performance is higher than the others in the encoder. Furthermore, by transmitting the representation coefficients of the removed MBs, these MBs can be accurately reconstructed in the decoder. Since the reconstruction performance is better than that of the conventional approach, the proposed method can remove more MBs from the target video sequences, and reduction of total amount of bits can be feasible. Therefore, the proposed method realizes the improvement of the video coding efficiency. Some experimental results are shown to verify the superior performance of the proposed method to that of H.264/AVC. The results also demonstrate that the bit-saving performance of the proposed method is comparable to that of H.2651 HEVC. (C) 2015 Elsevier B.V. All rights reserved.
  • Player Tracking in Far-View Soccer Videos Based on Composite Energy Function
    Kazuya Iwai, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, E97D, 7, 1885, 1892, IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG, 2014年07月, [査読有り]
    英語, 研究論文(学術雑誌), In this paper, an accurate player tracking method in far-view soccer videos based on a composite energy function is presented. In far-view soccer videos, player tracking methods that perform processing based only on visual features cannot accurately track players since each player region becomes small, and video coding causes color bleeding between player regions and the soccer field. In order to solve this problem, the proposed method performs player tracking on the basis of the following three elements. First, we utilize visual features based on uniform colors and player shapes. Second, since soccer players play in such a way as to maintain a formation, which is a positional pattern of players, we use this characteristic for player tracking. Third, since the movement direction of each player tends to change smoothly in successive frames of soccer videos, we also focus on this characteristic. Then we adopt three energies: a potential energy based on visual features, an elastic energy based on formations and a movement direction-based energy. Finally, we define a composite energy function that consists of the above three energies and track players by minimizing this energy function. Consequently, the proposed method achieves accurate player tracking in far-view soccer videos.
  • A new method for error degree estimation in numerical weather prediction via MKDA-based ordinal regression
    Takahiro Ogawa, Shintaro Takahashi, Sho Takahashi, Miki Haseyama
    EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING, 2014, 1, 115, 115, SPRINGER INTERNATIONAL PUBLISHING AG, 2014年07月, [査読有り]
    英語, 研究論文(学術雑誌), This paper presents a new method for estimating error degrees in numerical weather prediction via multiple kernel discriminant analysis (MKDA)-based ordinal regression. The proposed method tries to estimate how large prediction errors will occur in each area from known observed data. Therefore, ordinal regression based on KDA is used for estimating the prediction error degrees. Furthermore, the following points are introduced into the proposed approach. Since several meteorological elements are related to each other based on atmospheric movements, the proposed method merges such heterogeneous features in the target and neighboring areas based on a multiple kernel algorithm. This approach is based on the characteristics of actual meteorological data. Then, MKDA-based ordinal regression for estimating the prediction error degree of a target meteorological element in each area becomes feasible. Since the amount of training data obtained from known observed data becomes very large in the training stage of MKDA, the proposed method performs simple sampling of those training data to reduce the number of samples. We effectively use the remaining training data for determining the parameters of MKDA to realize successful estimation of the prediction error degree.
  • Bayesian Network-based Distress Estimation Using Image Features in Road Structure Assessment
    Keisuke Maeda, Sho Takahashi, Takahiro Ogawa, Miki Haseyama
    2014 IEEE 3RD GLOBAL CONFERENCE ON CONSUMER ELECTRONICS (GCCE), 169, 170, IEEE, 2014年, [査読有り]
    英語, 研究論文(国際会議プロシーディングス), This paper presents a Bayesian network-based method for estimating a distress of road structures from inspection data. The distress is represented by a damage of road structures and its degree. In the previous work, the distress was estimated by utilizing Bayesian network based on categories of road structures, details of road structures and damaged parts. However, inspection data include not only the above items but also images of the distress. Therefore, by introducing the use of the images to the previous work, improvement of the distress estimation accuracy can be expected. The proposed method calculates Bayesian network from inspection items and their corresponding images to perform the distress estimation. Experimental results show the effectiveness of the proposed method.
  • レベルセット法を用いたサッカー映像における選手追跡手法
    高橋 翔, 林 原局, 長谷山 美紀
    電子情報通信学会論文誌. D, 情報・システム = The IEICE transactions on information and systems (Japanese edition), 96, 3, 695, 703, 2013年03月, [査読有り]
    日本語
  • ACTIVE GRID-BASED METHOD FOR VISUALIZING PASS REGIONS IN SOCCER VIDEOS
    Sho Takahashi, Miki Haseyama
    ELECTRONIC PROCEEDINGS OF THE 2013 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO WORKSHOPS (ICMEW), 1, 6, IEEE, 2013年, [査読有り]
    英語, 研究論文(国際会議プロシーディングス), This paper presents a method for visualizing pass regions that have high probabilities of the pass succeeding from broadcast soccer videos. In soccer matches, players discover pass regions based on player position geometry and player velocities. Therefore, by using player position geometry and player velocities, which are obtained from a broadcast soccer video, we can visualize pass regions. The proposed method is realized by the following two steps. 1) Generation of new three-dimensional data (volume data) for analyzing pass regions, which are not visible. 2) Visualization of pass regions. In the first step, volume data are generated from player position geometry and player velocities. By generating the volume data, which express the player position geometry and the player velocities, analysis of invisible pass regions is enabled. In the second step, by applying Active grid to the generated volume data, pass regions are visualized. Specifically, lattice points of the Active grid converge to the pass regions. Therefore, positions of the pass regions on the pitch can be visualized from densities of the lattice points. In the experiment, the proposed method is applied to actual TV programs to verify its effectiveness.
  • Adaptive parameter setting for pass region estimation in soccer videos and its performance verification
    Sho Takahashi, Miki Haseyama
    2013 IEEE 2nd Global Conference on Consumer Electronics, GCCE 2013, 271, 272, IEEE, 2013年, [査読有り]
    英語, 研究論文(国際会議プロシーディングス), This paper proposes an accurate pass region estimation method by introducing adaptive parameter settings. Our previous paper proposed a pass region estimation method by utilizing average values of ball and player velocities. However, such velocities vary according to player density and skill. Therefore, in order to realize a more accurate pass region estimation, the proposed method obtains parameters, which are ball and player velocities, from player positions in a target soccer video. By introducing the above parameter settings to pass region estimation, more realistic pass region can be obtained. Consequently, the accurate method of pass region estimation is realized. © 2013 IEEE.
  • Active Grid-based Pass Region Estimation from Multiple Frames of Broadcast Soccer Videos
    Takahashi Sho, Haseyama Miki
    映像情報メディア学会英語論文誌, 1, 3, 220, 225, 一般社団法人 映像情報メディア学会, 2013年, [査読有り]
    英語, 研究論文(学術雑誌), An Active grid-based method for estimating pass regions from broadcast soccer videos is presented in this paper. It is assumed that the pass region has a high probability of the pass succeeding. In soccer matches, players discover pass regions based on previous and current player positions. In conventional methods, pass regions are estimated by applying Active Net to only a single frame of a soccer video. In the proposed method, Active grid is applied to three-dimensional data by which frames of the soccer video are connected with the temporal dimension. The proposed method then realizes robust estimation of pass regions based on multiple frames of player positions. The proposed method was applied to actual TV programs to verify its effectiveness.
  • アクティブネットを用いたサッカー映像におけるパス可能領域の推定
    高橋 翔, 今 宏史, 長谷山 美紀
    電子情報通信学会論文誌. D, 情報・システム = The IEICE transactions on information and systems (Japanese edition), 92, 4, 501, 510, 電子情報通信学会, 2009年04月01日, [査読有り]
    日本語, 研究論文(学術雑誌), 本論文では,チームスポーツ映像からアクティブネットを用いてパス可能領域を推定する手法を提案する.チームスポーツ映像の一つであるサッカー映像の意味内容解析を行うために重要なサッカーの戦術は,選手の移動とボール運びによって表現されるため,ボール運びを実現するパスを分析することは重要である.一般にパスコースはボール保持者と味方チームの選手へとつながる緩やかな曲線で表される.提案手法では,新たなエネルギーの定義とパス可能領域を推定するための画像生成により,アクティブネットを用いて前述の曲線が存在する領域を抽出する.また,パス可能領域は守備の選手から離れるほど,パスが成功する可能性が高いという特徴をもつ.提案手法では,格子点の密度に着眼することで,パスが成功する可能性をパス可能領域の推定と同時に得る.更に,アクティブネットの収束結果は多少の選手位置の誤差を許容するため,選手の動きを用いた従来手法における,選手位置の誤差の影響を受けやすいという問題点を解決することが可能である.したがって,提案手法はカメラワークが存在し,高精度な選手位置の推定が困難であるテレビ映像に対しても,高精度にパス可能領域の推定が可能である.

その他活動・業績

講演・口頭発表等

担当経験のある科目_授業

  • 応用多次元信号処理特論               
    北海道大学
    2020年12月 - 現在
  • 土木計画学               
    北海道大学
    2019年10月 - 現在
  • 土木計画学演習               
    北海道大学
    2019年04月 - 現在
  • 情報学Ⅰ               
    北海道大学
    2020年 - 2020年
  • 環境フィールド学実習               
    北海道大学
    2019年04月 - 2019年09月
  • 計算機演習               
    北海学園大学
    2009年04月 - 2011年03月

所属学協会

  • 自動車技術会               
  • 土木学会               
  • IEEE               
  • 映像情報メディア学会               
  • 電子情報通信学会               
  • 交通工学研究会               
  • 日本環境共生学会               

共同研究・競争的資金等の研究課題

  • 災害時の取得情報制約下における道路交通システムモニタリング
    科学研究費助成事業
    2023年04月01日 - 2026年03月31日
    杉浦 聡志, 高橋 翔, 倉内 文孝, 中西 航
    日本学術振興会, 基盤研究(B), 北海道大学, 23H01521
  • 積雪寒冷地のリアルタイムデジタルツインの構築および防災システムの創出
    科学研究費助成事業 基盤研究(B)
    2022年04月01日 - 2026年03月31日
    高橋 翔, 杉浦 聡志, 萩原 亨
    日本学術振興会, 基盤研究(B), 北海道大学, 22H01607
  • デジタルツインによる冬期道路交通マネジメントシステムの技術開発               
    道路政策の質の向上に資する技術研究開発
    2023年04月 - 2026年03月
    髙橋翔, 萩原亨, 有村幹治, 浅田拓海, 永田泰浩, 大井元揮, 芝崎拓, 小西信義, 丹治和博, 小松麻美, 槌本陽, 山本郁淳, 内藤利幸, 松田真宜, 高橋歩夢
    国土交通省, 本格研究, 新道路技術会議, 研究代表者
  • 冬期の自動運転を支援する道路管理システムに関する研究
    科学研究費助成事業 基盤研究(B)
    2019年04月01日 - 2023年03月31日
    萩原 亨, 宗廣 一徳, 高橋 翔, 有村 幹治
    本研究では交通シミュレーションソフトの VISSIMを用いて、冬期路面を想定したACC制御車の混在が低交通量の交通流に与える影響について明らかにする。最初に、ACC制御車の走行をVISSIMで再現する追従モデルを開発した。次に、フィールド実験におけるACC制御車(追従車)の追従挙動を用いてVISSIMで再現したACC制御車の追従モデルの妥当性を検証した。これらの結果をベースに、冬期路面を想定した長い設定車間時間の影響を検討した。交通流シミュレーションの結果から、ACC制御車の混在率が高く、設定車間時間が長いとき、平均速度・平均旅行時間・平均遅れ時間に大きな影響はない結果となった。ただし、ACC制御車の混在率が高く時間交通量が多い条件で、設定車間時間を長くすると平均遅れ時間が長くなるなど、交通流へ影響する結果となった。
    日本学術振興会, 基盤研究(B), 北海道大学, 19H02254
  • 冬期の自動運転を支援する道路管理システムに関する研究               
    科学研究費補助金(基盤研究(B))
    2019年04月 - 2023年03月
    萩原 亨
    文部科学省, 競争的資金
  • 次世代高精度検索を実現するスーパーマルチモーダル人間情報解析基盤
    科学研究費助成事業 基盤研究(B)
    2017年04月01日 - 2022年03月31日
    長谷山 美紀, 高橋 翔, 小川 貴弘, 畠山 泰貴
    【平成30年度の研究実施計画:異種データの関連性に基づき興味推定を高精度化する超グラフ解析技術の開発】
    平成30年度の研究では、マルチメディアデータとセンサーデータを同一の特徴空間で互いに比較可能とする超グラフの構築が可能となった。超グラフの解析により、共通の興味を有するユーザとコンテンツを統合的にグルーピングし、コミュニティとして抽出可能とすることで、興味に共通性を有する他ユーザやコンテンツを活用した、興味推定の高精度化を実現した。
    具体的には、マルチメディアデータとセンサーデータを同一の特徴空間で互いに比較可能とする超グラフを構築した後、ランダムウォークやHITSアルゴリズム等のグラフ解析法を高度化し、異種データ間の関連性を明らかにする新たな手法を実現した。これにより、共通の興味を有するコンテンツとユーザを統合的にグルーピングし、コミュニティとして抽出することが可能となり、ユーザとコンテンツを同時に活用した興味推定の高精度化が実現された。本年度には、最終年度の実証実験も視野に入れて、検索のための可視化インタフェースの設計に着手する。この際、検索の高精度化のための適合フィードバックを可能とするインタラクティブな可視化インタフェースを構築するとともに、各ユーザの行動履歴やセンサーデータの蓄積を行った。
    日本学術振興会, 基盤研究(B), 北海道大学, 17H01744
  • 次世代高精度検索を実現するスーパーマルチモーダル人間情報解析基盤               
    科学研究費補助金(基盤研究(B))
    2017年04月 - 2022年03月
    長谷山 美紀
    文部科学省, 競争的資金
  • 環境や利用者に適応して多様なスポーツのコンテンツを提示する次世代可視化技術の構築
    科学研究費助成事業 基盤研究(C)
    2017年04月01日 - 2021年03月31日
    高橋 翔
    スポーツの中でも,特に,複数人が1つのフィールド上で連携して試合を進めるサッカーやラグビー,野球等の競技では, フォーメーションが戦術の重要な要素であり,その分析は,試合内容を理解するために重要である.このため,提案者はこれまでにサッカーや野球の分析データを生成する理論の構築を進めており,本研究課題では,このさらなる高度化と多様化を進めるとともに,利用者のレベルや興味に応じて選手やチームの情報,戦況,さらには戦術の高度な情報までを動的に提示する次世代の可視化技術を構築する. すなわち,本研究では,利用者の「知識レベル」や「観戦・練習の環境」等の多様なデータから,「所望する分析データ」と「利用環境に適応した提示方式」を定める理論を導出し,スポーツの効率的な観戦や教育等を助ける情報提示の基盤技術を構築する.具体的に応募者は,本研究を4年間で計画しており,期間内に次の【フェーズ1】~【フェーズ4】を明らかにする.
    【フェーズ1】スポーツの分析データを生成する理論の高度化
    【フェーズ2】利用者の環境および操作履歴に基づく分析データ提示の基盤技術構築
    【フェーズ3】様々なスポーツ環境・デバイスでの分析データ提示システムの実装
    【フェーズ4】競技の枠を超えて適応的に分析データを提示する可視化技術への拡張
    平成30年度には,【フェーズ2】を特に進め,「操作履歴や観戦・練習等の利用環境のデータ」と「スポーツの分析データ」との相関を分析する理論を新たに導出し,適応的な分析データの可視化の基礎技術を構築した.具体的には,種々提案されている相関分析の手法や距離計量学習[8]を拡張する新たな手法を導出し,利用者の視線データや検索システムの操作と「分析データ」との関係を分析することで,利用者に適応した提示を可能とした.
    日本学術振興会, 基盤研究(C), 北海道大学, 17K00148
  • 自治体による観光情報発信支援のためのサイバーフィジカルデータ解析プラットフォームに関する研究開発               
    戦略的情報通信研究開発推進事業(SCOPE)
    2018年06月 - 2021年03月
    長谷山 美紀
    総務省, 競争的資金
  • 環境や利用者に適応して多様なスポーツのコンテンツを提示する次世代可視化技術の構築
    科学研究費補助金(基盤研究(C))
    2017年04月 - 2021年03月
    高橋 翔
    文部科学省, 研究代表者, 競争的資金
  • 個人の知識と経験に基づくスポーツコンテンツの動的な提示システムの構築
    科学研究費助成事業 若手研究(B)
    2014年04月01日 - 2018年03月31日
    高橋 翔
    本研究では,スポーツの観戦時などに有用なデータをユーザの経験と思考に基づいて提示するシステムを構築するため,放送映像やスタジアムで撮影した映像,インターネットなどで文字データとして公開される実況などのマルチメディアコンテンツとユーザの操作履歴から,パスコースやチーム間の優位度,自動生成される実況など,試合の理解に有用なデータを観戦する場所およびそれぞれのユーザに合わせて提示可能となる手法についての研究を進めた.
    日本学術振興会, 若手研究(B), 北海道大学, 研究代表者, 競争的資金, 26730057
  • 映像の意味理解を実現する映像解析手法の確立とその応用に関する研究
    科学研究費補助金(特別研究奨励費)
    2011年04月 - 2013年03月
    高橋 翔
    文部科学省, 研究代表者, 競争的資金
  • 映像の意味理解を実現する映像解析手法の確立とその応用に関する研究
    科学研究費助成事業 特別研究員奨励費
    2011年 - 2012年
    高橋 翔
    サッカー映像や野球映像などのスポーツ映像は世界中の人々が関心を持ち,非常に多くの視聴者が存在する.さらに,スポーツ映像はプロやアマチュアを問わずスポーツ教育へ利用されており,非常に重要な映像資料として扱われている.しかしながら,それらのスポーツ映像を視聴者が十分に理解するためには,選手や戦術などに関する知識や経験が要求され,視聴者が映像を視聴可能な時間は限られている.このため,映像を効果的に視聴し,内容を理解することを可能とする技術が必要とされている,そこで,申請者は映像を効果的に視聴可能とする技術の確立のために「高レベル特徴量を用いた映像意味理解に関する映像解析」および「映像解析手法のスポーツ教育への応用」について研究を進めた.
    平成24年度は,平成23年度において計画を達成することにより確立された映像意味理解に関する映像解析手法を用い,選手及びチームの技術向上や新たな技術の獲得を目的としたスポーツ教育への応用についての検討を行った,具体的には,平成23年度に実現された類似場面検索を行うことでテレビ放送や個人で撮影された大量のスポーツ映像の中から参考となるプレーや議論の対象となる場面の検索が可能となる手法を提案した.また,平成23年度の成果に基づき,擬似的な試合を効果的にシミュレーションすることを可能とした.以上のように,平成24年度は,映像意味理解に関する映像解析手法のスポーツ教育への応用についての検討を進めた.
    日本学術振興会, 特別研究員奨励費, 北海道大学, 11J01938
  • 深層学習によるAI機構を有するARを用いたデータ提示の基礎システム構築               
    研究助成金
    高橋 翔
    戸田育英財団, 研究代表者, 競争的資金

産業財産権

  • 路面状態判定装置、路面状態判定システム、車両、路面状態判定方法、及びプログラム               
    特許権, 岩崎悠志, 高橋翔, 萩原亨, 大廣智則, 株式会社ブリヂストン, 株式会社ネクスコ・エンジニアリング北海道
    特願2021-088688, 2021年05月26日
    特開2022-181641, 2022年12月08日
  • 路面状態判定装置、路面状態判定システム、車両、路面状態判定方法、及びプログラム               
    特許権, 岩崎悠志, 高橋翔, 萩原亨, 大廣智則, 株式会社ブリヂストン, 株式会社ネクスコ・エンジニアリング北海道
    2021-088683, 2021年05月26日
    特開2022-181640, 2022年12月08日
  • 体験記録システム、体験記録方法および体験記録プログラム
    特許権, 嶌田 聡, 東野 豪, 長谷山 美紀, 高橋 翔, 日本電信電話株式会社, 国立大学法人北海道大学
    特願2012-277957, 2012年12月20日
    特開2014-123817, 2014年07月03日
    特許第5920785号
    2016年04月22日
    201603007762337580
  • 映像アノテーション付与装置およびその動作方法
    特許権, 嶌田 聡, 長谷山 美紀, 高橋 翔, 日本電信電話株式会社, 国立大学法人北海道大学
    特願2011-152899, 2011年07月11日
    特開2013-021482, 2013年01月31日
    201303058263892263

担当教育組織