⚽

内田郁真（いくま）

A new look into Off-ball Scoring Opportunity: taking into account the continuous nature of the game

Machine LearningSports Evaluation Metrics

SportsTomorrow 2020

既存のOBSOはイベントデータに紐づくことが前提だったが、それがトラッキングデータと紐付けることが可能になった。

Hugo M. R. Rios-Neto, Wagner Meira Jr., Pedro O.S. Vaz de Melo

Tracking DataSports Event Data

OBSO：Beyond Expected Goals

Machine LearningSports Evaluation Metrics

MITSSAC 2018

典型的な指標はボール保持者における評価は取り組まれているが、ボール非保持の選手のプレー評価を定量的に行った研究は初めて

William Spearman

Tracking DataSports Event Data

https://www.notion.so/jsaasarp/TTNet-Real-time-temporal-and-spatial-video-analysis-of-table-tennis-e2ee8823a2e04d99ac24838a552212fc?pvs=4 https://www.notion.so/jsaasarp/Automated-Offside-Detection-by-Spatio-Temporal-Analysis-of-Football-Videos-04e34a049f5346daa02f2a7e03d732d1?pvs=4 https://www.notion.so/jsaasarp/A-new-look-into-Off-ball-Scoring-Opportunity-taking-into-account-the-continuous-nature-of-the-game-d155467d5aa543bab4a5d1ec6ffce935?pvs=4 https://www.notion.so/jsaasarp/OBSO-Beyond-Expected-Goals-d112572808fb4718b0509c93b1acfe95?pvs=4 https://www.notion.so/jsaasarp/Thinking-the-GOAT-Imitating-Tennis-Styles-6e94730e9f7a4a008f90ec4381fa6638?pvs=4 https://www.notion.so/jsaasarp/Fine-grained-retrieval-of-sports-plays-using-tree-based-alignment-of-trajectories-e78a10640d4448ac9517a5146072e7e6?pvs=4 https://www.notion.so/jsaasarp/Learning-to-track-and-identify-players-from-broadcast-sports-videos-e7a628a8d9074fe89576a575ac8f6c20?pvs=4 https://www.notion.so/jsaasarp/An-examination-of-expected-goals-and-shot-efficiency-in-soccer-ffcd7fa51f074072a5c517a949dcfaa5?pvs=4 https://www.notion.so/jsaasarp/Evaluation-of-soccer-players-to-create-scoring-opportunities-for-teammates-based-on-their-trajectory-7ecfb6de8fff435cb80b0f70ff0bd6a1?pvs=4 https://www.notion.so/jsaasarp/SIMPLE-ONLINE-AND-REALTIME-TRACKING-WITH-A-DEEP-ASSOCIATION-METRIC-22e7a611b47447619e6d35640e2427b6?pvs=4 https://www.notion.so/jsaasarp/SportsTrack-An-Innovative-Method-for-Tracking-Athletes-in-Sports-Scenes-40d5f000b5934f74a295f0b2fc658f39?pvs=4 https://www.notion.so/jsaasarp/Spotting-Temporally-Precise-Fine-Grained-Events-in-Video-f409d1a783794234afa314c20c863eaa?pvs=4 https://www.notion.so/jsaasarp/A-Survey-on-Video-Action-Recognition-in-Sports-Datasets-Methods-and-Applications-ebacb293e7d346c28e557f8d0823b0cd?pvs=4 https://www.notion.so/jsaasarp/Segment-Anything-b1499a2b4e494f7183587b1cbcfdd41b?pvs=4 https://www.notion.so/jsaasarp/Group-Activity-Recognition-in-Basketball-Tracking-Data-Neural-Embeddings-in-Team-Sports-NETS-fcfd926e9a4749beb0b2c74a0b93c46d?pvs=4 https://www.notion.so/jsaasarp/Human-Action-Recognition-without-Human-30951eb70cdd4c6f90b1f108b52b2fee?pvs=4 https://www.notion.so/jsaasarp/Diverse-Generation-for-Multi-agent-Sports-Games-b769b949bf1b4dfb9c37efce698f687d?pvs=4 https://www.notion.so/jsaasarp/c54e5627b592485696ac7b6e86272be6?pvs=4 https://www.notion.so/jsaasarp/88e023de52864ba8996529d7637c8f35?pvs=4 https://www.notion.so/jsaasarp/High-Speed-Tiny-Tennis-Ball-Detection-Based-on-Deep-Convolutional-Neural-Networks-05d25a6399aa46e086d7088b7fff33e2?pvs=4 https://www.notion.so/jsaasarp/Slicing-Aided-Hyper-Inference-and-Fine-tuning-for-Small-Object-Detection-85c68b8188e54fcda05302e073d64251?pvs=4 https://www.notion.so/jsaasarp/Contrastive-Learning-for-Sports-Video-Unsupervised-Player-Classification-dbbe99a4f5eb43fc83a2a6f66d48d68c?pvs=4 https://www.notion.so/jsaasarp/TrajGAT-A-Graph-based-Long-term-Dependency-Modeling-Approach-for-Trajectory-Similarity-Computation-aec4afc3d77540929aee977531c17785?pvs=4 https://www.notion.so/jsaasarp/Jointly-Contrastive-Representation-Learning-on-Road-Network-and-Trajectory-fefb35feeb644ead83196a0f6eb30364?pvs=4 https://www.notion.so/jsaasarp/Trajectory-Similarity-Learning-with-Auxiliary-Supervision-and-Optimal-Matching-5dc531c526414e87b5bd51a590bbcacc?pvs=4 https://www.notion.so/jsaasarp/Spatio-Temporal-Trajectory-Similarity-Measures-A-Comprehensive-Survey-and-Quantitative-Study-57a66d20be18431eab8e17953e270484?pvs=4 https://www.notion.so/jsaasarp/A-Survey-on-Deep-Learning-based-Spatio-temporal-Action-Detection-3f81a80f645445d78b7846c9fcf32637?pvs=4 https://www.notion.so/jsaasarp/TriDet-Temporal-Action-Detection-with-Relative-Boundary-Modeling-452ac391e058403db96e06b198ed2736?pvs=4 https://www.notion.so/jsaasarp/Towards-Active-Learning-for-Action-Spotting-in-Association-Football-Videos-9a5975ecb78c4f02b6337bb7651fcd8e?pvs=4

High-Speed Tiny Tennis Ball Detection Based on Deep Convolutional Neural Networks

Arxiv

Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection

Deep LearningMachine Learning

Arxiv 2022

Fatih Cagatay Akyon, Sinan Onur Altinuc, Alptekin Temizel

Video

Contrastive Learning for Sports Video: Unsupervised Player Classification

Computer VisionMachine LearningDeep Learning

CVSports 2021

教師情報不要でチームの分類が可能である．

Maria Koshkina, Hemanth Pidaparthy, James H. Elder

Video

TrajGAT: A Graph-based Long-term Dependency Modeling Approach for Trajectory Similarity Computation

Deep LearningMachine Learning

KDD 2022

多くの既存の軌跡類似性計算手法は短期的な局所特徴に依存しており、長期的な動きのパターンを把握することが難しい。TrajGATはこの問題をグラフ注意メカニズムによって解決

Di Yao , Haonan Hu , Lun Du , Gao Cong , Shi Han , Jingping Bi

Tracking Data

Jointly Contrastive Representation Learning on Road Network and Trajectory

Deep LearningMachine Learning

Arxiv 2022

先行研究では主に道路ネットワークまたはトラジェクトリの一方に焦点を当てていたが、この研究では両方を統合することで、より完全な情報を提供

Zhenyu Mao, Ziyue Li, Dedong Li, Lei Bai, Rui Zhao

Tracking Data

Trajectory Similarity Learning with Auxiliary Supervision and Optimal Matching

Machine LearningDeep Learning

IJCAI 2022

Hanyuan Zhang , Xingyu Zhang , Qize Jiang , Baihua Zheng , Zhenbang Sun , Weiwei Sun , Changhu Wang Authors Info & Claims

Tracking Data

Spatio-Temporal Trajectory Similarity Measures: A Comprehensive Survey and Quantitative Study

A Survey on Deep Learning-based Spatio-temporal Action Detection

Machine LearningDeep LearningComputer Vision

Arxiv 2023

Peng Wang , Fanwei Zeng , and Yuntao Qian

Video

TriDet: Temporal Action Detection with Relative Boundary Modeling

Computer VisionDeep LearningMachine Learning

CVPR 2023

Dingfeng Shi, Yujie Zhong, Qiong Cao, Lin Ma, Jia Li, Dacheng Tao

Towards Active Learning for Action Spotting in Association Football Videos

Computer VisionMachine LearningDeep Learning

CVSports

Action Spottingタスクはいずれも教師あり学習で大量のサンプルが必要．この論文では同タスクでのActiveLearning手法を初めて提案

Silvio Giancola, Anthony Cioppa, Julia Georgieva, Johsan Billingham, Andreas Serner, Kerry Peek, Bernard Ghanem, Marc Van Droogenbroeck

Sports Event DataVideo

Hierarchical Consistent Contrastive Learning for Skeleton-Based Action Recognition with Growing Augmentations

MotionBERT: A Unified Perspective on Learning Human Motion Representations

ICCV 2023

Wentao Zhu, Xiaoxuan Ma Zhaoyang Liu, Libin Liu, Wayne Wu, Yizhou Wang

Detector-free weakly supervised group activity recognition.

Computer VisionMachine LearningDeep Learning

CVPR 2022

Dongkeun Kim Jinsung Lee Minsu Cho, Suha Kwak

VideoSports Event Data

HumanBench: Towards General Human-centric Perception with Projector Assisted Pretraining

CVPR 2023

Two-Stream Networks for Weakly-Supervised Temporal Action Localization with Semantic-Aware Mechanisms

Computer VisionMachine LearningDeep Learning

CVPR 2023

Yu Wang, Yadong Li , Hongbin Wang

Video

SPARTAN: Self-Supervised Spatiotemporal Transformers Approach to Group Activity Recognition

Computer VisionMachine LearningDeep Learning

CVPR 2023

Global Temporal ViewsとLocal Spatiotemporal Viewsという概念を導入し，フレームの集合（ビュー）と他のフレームとの関連度（類似度）を学習している点が新しい．

Naga VS Raviteja Chappa , Pha Nguyen, Alexander H Nelson , Han-Seok Seo , Xin Li Page Daniel Dobbs ,Khoa Luu

VideoSports Event Data

Actionlet-Dependent Contrastive Learning for Unsupervised Skeleton-Based Action Recognition

Computer VisionDeep LearningMachine Learning

CVPR 2023

この手法ではActionletを用いることで，アクション領域の非アクション領域のを切り離すための特徴量を構築できるところがユニーク

Lilang Lin, Jiahang Zhang, Jiaying Liu*

Video

On the Benefits of 3D Pose and Tracking for Human Action Recognition

Computer VisionDeep Learning

CVPR 2023

Jathushan Rajasegaran, Georgios Pavlakos , Angjoo Kanazawa , Christoph Feichtenhofer, Jitendra Malik

Human Pose DataVideo

A Method of Football‑Ofside Detection Using Multiple Cameras for an Automatic Linesman Assistance System

Computer Vision

SpringerML 2019

マルチアングルにセットアップに拡張することでオクルージョンの問題に対応．性能面においてエキスパート相当の検出性能を達成

Sirimamayvadee Siratanita · Kosin Chamnongthai· Mitsuji Muneyasu

Sports Event Data

Spatio-Temporal Action Detection Under Large Motion

Computer VisionMachine LearningDeep Learning

WACV 2023

TOI-Align（Track-of-Interest Align）を用いた特徴抽出を提案し、TAAD（Track Aware Action Detector）を提案

Gurkirt Singh Vasileios Choutas Suman Saha Fisher Yu

VideoSports Event Data

An offside soccer detection system using ontology and deep learning

Computer VisionDeep LearningMachine Learning

TURCOMAT 2022

画像情報だけでなく音声情報を用いている点が新しい．

Mohammed Yassine Kazi Tani, Lamia Fatiha Kazi Tani , Abdelghani Ghomar

VideoAudio DataSports Event Data

Use of deep learning in soccer videos analysis: survey

Computer VisionMachine LearningDeep Learning

SpringerML 2022

Sara Akan, Songül Varlı

Simple Cues Lead to a Strong Multi-Object Tracker

CVPR 2023

MotionTrack: Learning Robust Short-term and Long-term Motions for Multi-Object Tracking

Deep LearningComputer VisionMachine Learning

CVPR 2023

ターゲット間の相互作用をモデル化する「Interaction Module」と (3)識別可能なモーションパターンを学習する「Refind Module」を設計し、失われたトラックレットを現在の検出値で再識別できるようにしている。

Zheng Qin, Sanping Zhou, Le Wang, Jinghai Duan, Gang Hua, Wei Tang

VideoTracking Data

Iterative Scale-Up ExpansionIoU and Deep Features Association for Multi-Object Tracking in Sports

Computer VisionMachine LearningDeep Learning

Arxiv 2023

画像ベースのMOT手法と比較して、低い計算コストを維持しながら、追加の学習データを必要とせず、優れた追跡性能を達成

Hsiang-Wei Huang, Cheng-Yen Yang, Jenq-Neng Hwang, Chung-I Huang

VideoTracking Data

VARS: Video Assistant Referee System for Automated Soccer Decision Making From Multiple Views

Deep LearningMachine LearningComputer Vision

CVSports 2023

VARの自動化に動画認識手法を用いた初の取り組み

Jan Held, Anthony Cioppa, Silvio Giancola, Abdullah Hamdi, Bernard Ghanem, Marc Van Droogenbroeck

Video

Deep Representation Learning for Trajectory Similarity Computation

Deep LearningMachine Learning

ICDE 2018

Xiucheng Li , Kaiqi Zhao , Gao Cong, Christian S. Jensen , Wei Wei

Tracking Data

TrajGAT: A Graph-based Long-term Dependency Modeling Approach for Trajectory Similarity Computation

Deep LearningMachine Learning

KDD 2022

長期のシーケンスに対しての類似軌道が劇的に低下していた．trajGATにより長期記憶に対応

Di Yao , Haonan Hu , Lun Du , Gao Cong , Shi Han , Jingping Bi

Tracking Data

SportsPose – A Dynamic 3D sports pose dataset

Computer VisionDeep Learning

CVSports 2023

Human3.6Mや3DPWよりも多様で複雑な動作を含んだデータセットであり，またマーカーベースと比較して誤差が少ない，3DPWに匹敵する．

Christian Keilstrup Ingwersen , Christian Møller Mikkelstrup Janus Nørtoft Jensen, Morten Rieger Hannemose, Anders Bjorholm Dahl

Human Pose Data

Computing Trajectory Similarity in Linear Time: A Generic Seed-Guided Neural Metric Learning Approach

Arxiv 2019

NEUTRAJは、既存のあらゆる軌跡指標に対応する汎用性を持ち、与えられた軌跡ペアの類似性を線形時間で高速に計算する。さらに、 NEUTRAJは、空間ベースの軌跡インデックス作成手法と連携して、検索空間を縮小する．

Di Yao, Gao Cong, Chao Zhang, Jingping Bi

Tracking DataSimulation Data

A Boosted Model Ensembling Approach to Ball Action Spotting in Videos: The Runner-Up Solution to CVPR’23 SoccerNet Challenge

Deep LearningMachine LearningComputer Vision

Arxiv 2023

データセットのアンサンブル方法，学習時のデータのサンプリング方法に特に工夫が見られる．

Luping Wang, Hao Guo, Bin Liu

VideoSports Event Data

Ball Trajectory Inference from Multi-Agent Sports Contexts Using Set Transformer and Hierarchical Bi-LSTM

Deep LearningSports Evaluation Metrics

KDD 2023

今回は将来の軌道を予測する確率論的な問題ではなく，位置情報の予測ということでより決定論的なアプローチが求められる．この手法では復元タスクではなく回帰アプローチでタスクを解いている点は新しい．

Hyunsung Kim, Han-Jun Choi

Tracking Data

Deep Learning を用いた移動物体間のインタラクションを考慮した経路予測の研究動向

2022

Thinking the GOAT: Imitating Tennis Styles

MITSSAC 2023

Padmanaba Srinivasan, Raghavan Subramanian, William Knottenbelt,

Deep Representation Learning for Trajectory Similarity Computation

Machine LearningDeep Learning

Arxiv 2018

Xiucheng Li, Kaiqi Zhao , Gao Cong , Christian S. Jensen , Wei Wei

Tracking Data

Diverse Generation for Multi-agent Sports Games

Machine LearningDeep Learning

CVPR 2019

GNNと変分リカレントNN（VRNN）を活用しスポーツに適した順列等変量モデルを実現している点．

Raymond A. Yeh; Alexander G. Schwing; Jonathan Huang; Kevin Murphy

Tracking Data

Fine-grained retrieval of sports plays using tree-based alignment of trajectories

Machine LearningDeep Learning

Arxiv 2017

Long Sha, Patrick Lucey, Stephan Zheng, Taehwan Kim, Yisong Yue, Sridha Sridharan

Tracking Data

Group Activity Recognition in Basketball Tracking Data - Neural Embeddings in Team Sports (NETS)

Arxiv 2022

Sandro Hauri, Slobodan Vucetic

Tracking Data

Human Action Recognition without Human

Computer VisionDeep LearningMachine Learning

Arxiv 2016

この研究では人間を含まない背景のみを分析することで行動認識が可能かどうかを検証している。

Yun He, Soma Shirakabe, Yutaka Satoh, Hirokatsu Kataoka

Video

Segment Anything

Computer VisionDeep LearningMachine Learning

Arxiv 2023

Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C. Berg, Wan-Yen Lo, Piotr Dollár, Ross Girshick

VideoSegmentation Data

A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications

Computer VisionMachine LearningDeep Learning

Arxiv 2022

スポーツのドメインを包括的に調べたペーパーはおそらく初

Fei Wu, Qingzhong Wang, Jian Bian, Haoyi Xiong, Ning Ding, Feixiang Lu, Jun Cheng, Dejing Dou

Sports Event DataVideo

Spotting Temporally Precise, Fine-Grained Events in Video

Computer VisionMachine LearningDeep Learning

ECCV 2022

テニスとフィギアスケートの動画認識タスクにおいてSOTA

James Hong, Haotian Zhang, Michaël Gharbi, Matthew Fisher, Kayvon Fatahalian

Sports Event DataVideo

TTNet: Real-time temporal and spatial video analysis of table tennis

Computer VisionDeep LearningMachine Learning

CVSports 2020

Roman Voeikov, Nikolay Falaleev, Ruslan Baikulov

VideoSports Event Data

Automated Offside Detection by Spatio-Temporal Analysis of Football Videos

Computer VisionDeep LearningMachine Learning

MMSports 2021

Ikuma Uchida, Atom Scott, Hidehiko Shishido, Yoshinari Kameda

VideoSports Event Data

SIMPLE ONLINE AND REALTIME TRACKING WITH A DEEP ASSOCIATION METRIC

Computer VisionDeep LearningMachine Learning

Arxiv 2017

SORTのパフォーマンス向上のために、外観情報を統合している。

Nicolai Wojke, Alex Bewley, Dietrich Paulus

VideoTracking Data

SportsTrack: An Innovative Method for Tracking Athletes in Sports Scenes

Deep LearningMachine Learning

Arxiv 2022

SportsMOTチャレンジ一位の解法。

Jie Wang, Yuzhou Peng, Xiaodong Yang, Ting Wang, Yanming Zhang

VideoTracking Data

Learning to track and identify players from broadcast sports videos

Computer VisionDeep LearningMachine Learning

TPAMI 2013

シングルカメラのスポーツ映像で選手の識別を行った点がユニーク

W.-L. Lu, J.-A. Ting, J. Little, K. Murphy

Tracking DataVideo

An examination of expected goals and shot efficiency in soccer

Sports Evaluation Metrics

Arxiv 2016

複数のゴール期待値の研究をサーベイし、またxGの価値と信頼性を学術的に明らかにした点

Alex Rathke

Sports Event Data

Evaluation of soccer players to create scoring opportunities for teammates based on their trajectory prediction

Sports Evaluation MetricsDeep Learning

MLSA 2022

これまでのボール非保持選手の評価指標と比べて、「ボールを持たず、ボールを受けようとしない攻撃選手」を評価し、典型的な（あるいは予測された）動きと比較して、動きが得点機会の創出にどのように寄与するかを明らかにすることはが可能となった。

Masakiyo Teranishi, Kazushi Tsutsui, Kazuya Takeda, Keisuke Fujii

Tracking DataSports Event Data

Comparing cueing and constraints interventions for increasing launch angle in baseball batting.

Sports Evaluation Metrics

SoccerTrack: A Dataset and Tracking Algorithm for Soccer with Fish-eye and Drone Videos

Computer Vision

Pose Tutor: An Explainable System for Pose Correction in the Wild

Computer VisionDeep Learning

SoccerNet-Tracking: Multiple Object Tracking Dataset and Benchmark in Soccer Videos

Computer VisionDeep Learning

Pass Receiver Prediction in Soccer using Video and Players’ Trajectories

Computer VisionDeep Learning

Actions Speak Louder than Goals: Valuing Player Actions in Soccer

Machine LearningSports Evaluation Metrics

A framework for the fine-grained evaluation of the instantaneous expected value of soccer possessions

Sports Evaluation MetricsMachine Learning

A new look into Off-ball Scoring Opportunity: taking into account the continuous nature of the game

Machine LearningSports Evaluation Metrics

OBSO：Beyond Expected Goals

Machine LearningSports Evaluation Metrics

Diverse Generation for Multi-agent Sports Games

Machine LearningDeep Learning

Spotting Temporally Precise, Fine-Grained Events in Video

Computer VisionMachine LearningDeep Learning

SportsTrack: An Innovative Method for Tracking Athletes in Sports Scenes

Deep LearningMachine Learning

Self-Supervised Learning for Human Pose Estimation in Sports

Deep LearningComputer Vision

Learning to track and identify players from broadcast sports videos

Computer VisionDeep LearningMachine Learning

An examination of expected goals and shot efficiency in soccer

Sports Evaluation Metrics

Evaluation of soccer players to create scoring opportunities for teammates based on their trajectory prediction

Sports Evaluation MetricsDeep Learning