見出し画像

CVSports 2024に参加しました

Keisuke Fujii (藤井慶輔)

パターン認識とコンピュータビジョンのトップ国際会議CVPRにて開催されている、スポーツ領域に特化したワークショップCVSportsに参加してきました。
詳しくはフットボリスタの記事に書いたのですが、ここでは、CVSports 2024で発表された24本の論文のリンクと公開されていたコードを共有します（私の研究室からも3件の発表が行われ、全て公開しています）。

様々なトピックがありますが、スポーツアナリティクスに重要そうな観点から以下に整理しました（CVSportsの位置づけについてはこちら）。

  • Field registration & Calibration

    • No Bells Just Whistles: Sports Field Registration by Leveraging Geometric Properties Video Interaction Recognition using an Attention Augmented Relational Network and Skeleton Data, Marc Gutiérrez-Pérez, Antonio Agudo [pdf][code]

    • A Universal Protocol to Benchmark Camera Calibration for Sports, Floriane Magera, Thomas Hoyoux, Olivier Barnich, Marc Van Droogenbroeck [pdf]

  • Tracking

    • TeamTrack: A Dataset for Multi-Sport Multi-Object Tracking in Full-pitch Videos, Atom Scott, Ikuma Uchida, Ning Ding, Rikuhei Umemoto, Rory Bunker, Ren Kobayashi, Takeshi Koyama, Masaki Onishi, Yoshinari Kameda, Keisuke Fujii [pdf] [code]

    • MV-Soccer: Motion-Vector Augmented Instance Segmentation for Soccer Player Tracking, Fahad Majeed, Nauman Ullah Gilal, Khaled Al-Thelaya, Yin Yang, Marco Agus, Jens Schneider [pdf] 

  • Re-ID (Jersey number recognition)

    • A General Framework for Jersey Number Recognition in Sports Video, Maria Koshkina, James H. Elder [pdf][code] 

  • Pose estimation

    • Pseudo-label Based Unsupervised Fine-tuning of a Monocular 3D Pose Estimation Model for Sports Motions, Tomohiro Suzuki, Ryota Tanaka, Kazuya Takeda, Keisuke Fujii [pdf] [code]

  • Scene understanding including action recognition and detection

    • Video Interaction Recognition using an Attention Augmented Relational Network and Skeleton Data, Farzaneh Askari, Cyril Yared, Rohit Ramaprasad, Devin Garg, Anjun Hu, James J. Clark [pdf](Hockey)

    • Beyond the Premier: Assessing Action Spotting Transfer Capability Across Diverse Domains, Bruno Cabado, Anthony Cioppa, Silvio Giancola, Andrés Villa, Bertha Guijarro-Berdiñas, Emilio J. Padrón, Bernard Ghanem, Marc Van Droogenbroeck [pdf]  (examine many datasets)

    • Medium Scale Benchmark for Cricket Excited Actions Understanding, Altaf Hussain, Noman Khan, Muhammad Munsif, Min Je Kim, Sung Wook Baik [pdf]

    • Rugby Scene Classification Enhanced by Vision Language Model, Naoki Nonaka, Ryo Fujihira, Toshiki Koshiba, Akira Maeda, Jun Seita [pdf]

    • T-DEED: Temporal-Discriminability Enhancer Encoder-Decoder for Precise Event Spotting in Sports Videos, Artur Xarles, Sergio Escalera, Thomas B. Moeslund, Albert Clapés [pdf] [code] (1st place in ball action spotting 2024)

  • Integration of the above

    • AutoSoccerPose: Automated 3D Posture Analysis of Soccer Shot Movements, Calvin Yeung, Kenjiro Ide, Keisuke Fujii [pdf] [code] 

    • SoccerNet Game State Reconstruction: End-to-End Athlete Tracking and Identification on a Minimap, Vladimir Somers, Victor Joos, Anthony Cioppa, Silvio Giancola, Seyed Abolfazl Ghasemzadeh, Floriane Magera, Baptiste Standaert, Amir M. Mansourian, Xin Zhou, Shohreh Kasaei, Bernard Ghanem, Alexandre Alahi, Marc Van Droogenbroeck, Christophe De Vleeschouwer [pdf] [code]

    • Hierarchical NeuroSymbolic Approach for Comprehensive and Explainable Action Quality Assessment, Lauren Okamoto, Paritosh Parmar [pdf] [supp] [code]

  • Utilization of simulation datasets

    • Augmenting Pass Prediction via Imitation Learning in Soccer Simulations, Takeshi Kaneko, Rei Kawakami, Takeshi Naemura, Nakamasa Inoue [pdf] [supp] 

    • SoccerNet-Depth: a Scalable Dataset for Monocular Depth Estimation in Sports Videos, Arnaud Leduc, Anthony Cioppa, Silvio Giancola, Bernard Ghanem, Marc Van Droogenbroeck [pdf][code]

  • Action prediction and evaluation

    • PitcherNet: Powering the Moneyball Evolution in Baseball Video Analytics, Jerrin Bright, Bavesh Balaji, Yuhao Chen, David A Clausi, John S Zelek [pdf] 

    • A Stroke of Genius: Predicting the Next Move in Badminton, Magnus Ibh, Stella Graßhof, Dan Witzner Hansen [pdf]

    • Multi-Modal Hit Detection and Positional Analysis in Padel Competitions, Robbe Decorte, Martin Paré, Jelle Vanhaeverbeke, Joachim Taelman, Maarten Slembrouck, Steven Verstockt [pdf][data]
      Other topics

  • Ball

    • Event-based Ball Spin Estimation in Sports, Takuya Nakabayashi, Kyota Higa, Masahiro Yamaguchi, Ryo Fujiwara, Hideo Saito [pdf]

    • Table Tennis Ball Spin Estimation with an Event Camera, Thomas Gossard, Julian Krismer, Andreas Ziegler, Jonas Tebbe, Andreas Zell [pdf]

  • Integration with language

    • FineRehab: A Multi-modality and Multi-task Dataset for Rehabilitation Analysis, Jianwei Li, Jun Xue, Rui Cao, Xiaoxia Du, Siyu Mo, Kehao Ran, Zeyan Zhang [pdf]

    • ExerAIde: AI-assisted Multimodal Diagnosis for Enhanced Sports Performance and Personalised Rehabilitation, Ahmed Qazi, Asim Iqbal [pdf] 

    • X-VARS: Introducing Explainability in Football Refereeing with Multi-Modal Large Language Models, Jan Held, Hani Itani, Anthony Cioppa, Silvio Giancola, Bernard Ghanem, Marc Van Droogenbroeck [pdf] [code]

コンピュータビジョンの研究は多すぎて追いつけてなかったのですが、飛行機の中で論文を読んで、色々な研究者と話して、どういう研究が評価されるのか、自分たちがどういう方向に進むべきなのか、何となく見えた気がしました。また機会があればお話しできたらと思います。

この記事が気に入ったらサポートをしてみませんか？