Faculty Profiles - FUJII Keisuke

Papers - FUJII Keisuke

Division display 21 - 40 of about 105 ／ All the affair displays >>

Evaluating soccer match prediction models: a deep learning approach and feature optimization for gradient-boosted trees. Reviewed

Calvin Yeung, Rory P. Bunker, Rikuhei Umemoto, Keisuke Fujii

Machine Learning Vol. 113 ( 10 ) page： 7541 - 7564 2024.10

3D Pose-Based Temporal Action Segmentation for Figure Skating: A Fine-Grained and Jump Procedure-Aware Annotation Approach. Reviewed

Ryota Tanaka, Tomohiro Suzuki, Keisuke Fujii

Proceedings of the 7th ACM International Workshop on Multimedia Content Analysis in Sports(MMSports@MM) page： 17 - 26 2024.10

Multi-agent statistically discriminative sub-trajectory mining and an application to NBA basketball Reviewed

Rory Paul Bunker, Vo Nguyen Le Duy, Yasuo Tabei, Ichiro Takeuchi, Keisuke Fujii

Journal of Quantitative Analysis in Sports 2024.9

Authorship：Last author,　Corresponding author Language：English Publishing type：Research paper (scientific journal) Publisher：Walter de Gruyter GmbH

Abstract

Improvements in tracking technology through optical and computer vision systems have enabled a greater understanding of the movement-based behaviour of multiple agents, including in team sports. In this study, a multi-agent statistically discriminative sub-trajectory mining (MA-Stat-DSM) method is proposed that takes a set of binary-labelled agent trajectory matrices as input and incorporates Hausdorff distance to identify sub-matrices that statistically significantly discriminate between the two groups of labelled trajectory matrices. Utilizing 2015/16 SportVU NBA tracking data, agent trajectory matrices representing attacks consisting of the trajectories of five agents (the ball, shooter, last passer, shooter defender, and last passer defender), were truncated to correspond to the time interval following the receipt of the ball by the last passer, and labelled as effective or ineffective based on a definition of attack effectiveness that we devise in the current study. After identifying appropriate parameters for MA-Stat-DSM by iteratively applying it to all matches involving the two top- and two bottom-placed teams from the 2015/16 NBA season, the method was then applied to selected matches and could identify and visualize the portions of plays, e.g., involving passing, on-, and/or off-the-ball movements, which were most relevant in rendering attacks effective or ineffective.

DOI： 10.1515/jqas-2023-0039

researchmap

Other Link： https://www.degruyter.com/document/doi/10.1515/jqas-2023-0039/pdf

Mathematical models for off-ball scoring prediction in basketball Reviewed

Rikako Kono, Keisuke Fujii

11th Workshop on Machine Learning and Data Mining for Sports Analytics 2024 (MLSA'24) co-located with the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML-PKDD'24) 2024.9

Authorship：Last author,　Corresponding author Language：English Publishing type：Research paper (international conference proceedings)

In professional basketball, the accurate prediction of scoring opportunities
based on strategic decision-making is crucial for spatial and player
evaluations. However, traditional models often face challenges in accounting
for the complexities of off-ball movements, which are essential for
comprehensive performance evaluations. In this study, we propose two
mathematical models to predict off-ball scoring opportunities in basketball,
considering pass-to-score and dribble-to-score sequences: the Ball Movement for
Off-ball Scoring (BMOS) and the Ball Intercept and Movement for Off-ball
Scoring (BIMOS) models. The BMOS model adapts principles from the Off-Ball
Scoring Opportunities (OBSO) model, originally designed for soccer, to
basketball, whereas the BIMOS model also incorporates the likelihood of
interception during ball movements. We evaluated these models using player
tracking data from 630 NBA games in the 2015-2016 regular season, demonstrating
that the BIMOS model outperforms the BMOS model in terms of team scoring
prediction accuracy, while also highlighting its potential for further
development. Overall, the BIMOS model provides valuable insights for tactical
analysis and player evaluation in basketball.

arXiv

researchmap

Other Link： http://arxiv.org/pdf/2406.08749v2

Force-Based Modeling of Heterogeneous Roles in the Coordinated Behavior of a Triad Reviewed

Jun Ichikawa, Keisuke Fujii

New Generation Computing Vol. 42 ( 5 ) page： 983 - 995 2024.8

Authorship：Last author Language：English Publishing type：Research paper (scientific journal) Publisher：Springer Science and Business Media LLC

Abstract

Group coordination is defined as interactions with other members to implement a task that is difficult to do alone or to achieve higher performance than an individual. Meanwhile, the adjustment process in coordination is not uniquely determined because each individual has many degrees of freedom. It is more difficult to explain and model complex and dynamic coordination, such as nonverbal behavior of three or more members than pair or verbal interaction. Hence, we previously introduced a coordinated drawing task and conducted the behavioral experiment. The triads operated reels to change the tensions of threads connected to a pen, shared three heterogeneous roles (pulling, relaxing, and adjusting), and moved the pen to draw an equilateral triangle. The results indicated that the adjusting role was related to high task performance by helping resiliently without disturbing the pen’s smooth movement while avoiding great pen deviation. However, this experiment alone cannot explain details of the adjustment process of tension. To supplement these findings, this study formulated the three roles using equations of motion. The multi-agent simulation results showed that the adjusting role might use the degree of pen deviation reflected by the others’ motor information, such as the operating procedures and forces, to change the tension and draw at least three sides. Although it is necessary to consider that we used the experimental task, our study contributes to the fundamental understanding of resilient adjustment in coordination by sharing heterogeneous roles as the first step.

DOI： 10.1007/s00354-024-00277-y

researchmap

Other Link： https://link.springer.com/article/10.1007/s00354-024-00277-y/fulltext.html

Open-Vocabulary Predictive World Models from Sensor Observations. Reviewed International journal

Robin Karlsson, Ruslan Asfandiyarov, Alexander Carballo, Keisuke Fujii, Kento Ohtani, Kazuya Takeda

Sensors Vol. 24 ( 14 ) page： 4735 - 4735 2024.7

Language：English Publishing type：Research paper (scientific journal)

Cognitive scientists believe that adaptable intelligent agents like humans perform spatial reasoning tasks by learned causal mental simulation. The problem of learning these simulations is called predictive world modeling. We present the first framework for a learning open-vocabulary predictive world model (OV-PWM) from sensor observations. The model is implemented through a hierarchical variational autoencoder (HVAE) capable of predicting diverse and accurate fully observed environments from accumulated partial observations. We show that the OV-PWM can model high-dimensional embedding maps of latent compositional embeddings representing sets of overlapping semantics inferable by sufficient similarity inference. The OV-PWM simplifies the prior two-stage closed-set PWM approach to the single-stage end-to-end learning method. CARLA simulator experiments show that the OV-PWM can learn compact latent representations and generate diverse and accurate worlds with fine details like road markings, achieving 69 mIoU over six query semantics on an urban evaluation sequence. We propose the OV-PWM as a versatile continual learning paradigm for providing spatio-semantic memory and learned internal simulation capabilities to future general-purpose mobile robots.

DOI： 10.3390/s24144735

PubMed

researchmap

Unveiling Multi-Agent Strategies: A Data-Driven Approach for Extracting and Evaluating Team Tactics from Football Event and Freeze-Frame Data. Reviewed

Calvin Yeung 0001, Rory P. Bunker, Keisuke Fujii 0001

Journal of Robotics and Mechatronics Vol. 36 ( 3 ) page： 603 - 617 2024.6

Authorship：Last author,　Corresponding author Language：English Publishing type：Research paper (scientific journal) Publisher：Fuji Technology Press Ltd.

Studying collective behavior in opposing multi-agent teams is crucial across game theory, robotics, and sports analytics. In sports, especially football, team tactics involve intricate strategic spatial and action behaviors displayed as event sequences during possession. Understanding and analyzing these tactics is essential for successful training, strategic planning, and on-field success. While traditional approaches, such as notational and statistical analyses, offer valuable insights into team tactics, they often lack a comprehensive consideration of contextual information, thereby limiting the holistic evaluation of teams’ performances. To bridge this gap and capture the nuanced intricacies of team tactics, we employed advanced methodologies. The sequential pattern mining algorithm PrefixSpan was utilized to extract tactical patterns from possession sequences, enabling a deeper understanding of how teams strategize and adapt during play. Additionally, the neural marked spatio temporal point process (NMSTPP) model was leveraged to model and predict team behaviors, facilitating a fair comparison among teams. The evaluation of team possessions was further enhanced through the innovative holistic possession utilization score metrics, providing a more nuanced assessment of performance. In our experimental exploration, we identified and classified five distinct team tactics, validated the efficacy of the NMSTPP model when integrating StatsBomb 360 data, and conducted a comprehensive analysis of English Premier League teams during the 2022/2023 season. The results were visualized using radar plots and scatter plots with mean shift clustering. Lastly, the potential applications to RoboCup were discussed.

DOI： 10.20965/jrm.2024.p0603

CiNii Research

researchmap

Force-Based Modeling of a Resilient Helping Role in Coordinated Behavior of a Triad. Reviewed

Jun Ichikawa, Keisuke Fujii

26th International Conference on Human-Computer Interaction (HCI International 2024) Vol. 56 page： 148 - 155 2024.6

AutoSoccerPose: Automated 3D posture Analysis of Soccer Shot Movements. Reviewed

Calvin Yeung, Kenjiro Ide, Keisuke Fujii

IEEE/CVF Conference on Computer Vision and Pattern Recognition page： 3214 - 3224 2024.6

Pseudo-label based unsupervised fine-tuning of a monocular 3D pose estimation model for sports motions. Reviewed

Tomohiro Suzuki, Ryota Tanaka, Kazuya Takeda, Keisuke Fujii

IEEE/CVF Conference on Computer Vision and Pattern Recognition page： 3315 - 3324 2024.6

TeamTrack: A Dataset for Multi-Sport Multi-Object Tracking in Full-pitch Videos. Reviewed

Atom Sco, Ikuma Uchida, Ning Ding, Rikuhei Umemoto, Rory P. Bunker, Ren Kobayashi, Takeshi Koyama, Masaki Onishi, Yoshinari Kameda, Keisuke Fujii

IEEE/CVF Conference on Computer Vision and Pattern Recognition page： 3357 - 3366 2024.6

Collaborative hunting in artificial agents with deep reinforcement learning Reviewed

Kazushi Tsutsui, Ryoya Tanaka, Kazuya Takeda, Keisuke Fujii

eLife 2024.5

Authorship：Last author Language：English Publishing type：Research paper (scientific journal) Publisher：Cold Spring Harbor Laboratory

ABSTRACT

Collaborative hunting, in which predators play different and complementary roles to capture prey, has been traditionally believed as an advanced hunting strategy requiring large brains that involve high level cognition. However, recent findings that collaborative hunting have also been documented in smaller-brained vertebrates have placed this previous belief under strain. Here, we demonstrate that decisions underlying collaborative hunts do not necessarily rely on sophisticated cognitive processes using computational multi-agent simulation based on deep reinforcement learning. We found that apparently elaborate coordination can be achieved through a relatively simple decision process of mapping between observations and actions via distance-dependent internal representations formed by prior experience. Furthermore, we confirmed that this decision rule of predators is robust against unknown prey controlled by humans. Our results of computational ecology emphasize that collaborative hunting can emerge in various intra- and inter-specific interactions in nature, and provide insights into the evolution of sociality.

DOI： 10.1101/2022.10.10.511517

researchmap

Estimation of control area in badminton doubles with pose information from top and back view drone videos. Reviewed

Ning Ding, Kazuya Takeda, Wenhui Jin, Yingjiu Bei, Keisuke Fujii 0001

Multimedia Tools and Applications Vol. 83 ( 8 ) page： 24777 - 24793 2024.5

Language：English Publishing type：Research paper (scientific journal) Publisher：Springer Science and Business Media LLC

Abstract

The application of visual tracking to the performance analysis of sports players in dynamic competitions is vital for effective coaching. In doubles matches, coordinated positioning is crucial for maintaining control of the court and minimizing opponents’ scoring opportunities. The analysis of such teamwork plays a vital role in understanding the dynamics of the game. However, previous studies have primarily focused on analyzing and assessing singles players without considering occlusion in broadcast videos. These studies have relied on discrete representations, which involve the analysis and representation of specific actions (e.g., strokes) or events that occur during the game while overlooking the meaningful spatial distribution. In this work, we present the first annotated drone dataset from top and back views in badminton doubles and propose a framework to estimate the control area probability map, which can be used to evaluate teamwork performance. We present an efficient framework of deep neural networks that enables the calculation of full probability surfaces. This framework utilizes the embedding of a Gaussian mixture map of players’ positions and employs graph convolution on their poses. In the experiment, we verify our approach by comparing various baselines and discovering the correlations between the score and control area. Additionally, we propose a practical application for assessing optimal positioning to provide instructions during a game. Our approach offers both visual and quantitative evaluations of players’ movements, thereby providing valuable insights into doubles teamwork. The dataset and related project code is available at https://github.com/Ning-D/Drone_BD_ControlArea

DOI： 10.1007/s11042-023-16362-1

researchmap

Other Link： https://link.springer.com/article/10.1007/s11042-023-16362-1/fulltext.html

DRUformer: Enhancing Driving Scene Important Object Detection With Driving Scene Relationship Understanding. Reviewed

Yingjie Niu, Ming Ding 0002, Keisuke Fujii 0001, Kento Ohtani, Alexander Carballo, Kazuya Takeda

IEEE Access Vol. 12 page： 67589 - 67599 2024.5

Automatic detection of faults in race walking from a smartphone camera: a comparison of an Olympic medalist and university athletes Reviewed

Tomohiro Suzuki, Kazuya Takeda, Keisuke Fujii

International Journal of Computer Science in Sport 2024.3

Automatic Detection of Faults in Simulated Race Walking from a Fixed Smartphone Camera. Reviewed

Tomohiro Suzuki, Kazuya Takeda, Keisuke Fujii 0001

International Journal of Computer Science in Sport Vol. 23 ( 1 ) page： 22 - 36 2024.3

Hierarchical Integration of Deep Reinforcement Learning with a Pursuit Behavioral Model for Robust and Interpretable Navigation

TSUTSUI Kazushi, TAKEDA Kazuya, FUJII Keisuke

Proceedings of the Annual Conference of JSAI Vol. JSAI2024 ( 0 ) page： 3I5OS27b04 - 3I5OS27b04 2024

Authorship：Last author Language：Japanese Publishing type：Research paper (scientific journal) Publisher：The Japanese Society for Artificial Intelligence

Integrating theoretical models into machine learning models holds immense potential for constructing efficient, robust, and interpretable models. Here, we propose a hybrid architecture that hierarchically integrates a biological pursuit model into deep reinforcement learning. This approach enables seamless acceleration-mode switching and geometrically reasonable action selection, demonstrating our hierarchical predator agents realized efficient navigation in a predator-prey environment. Interestingly, our results have commonalities with group hunting behaviors observed in nature, suggesting the potential application of our model as a tool for providing new insights into biology.

DOI： 10.11517/pjsai.jsai2024.0_3i5os27b04

CiNii Research

Space evaluation based on pitch control using drone video in Ultimate

IWASHITA Shunsuke, SCOTT Atom, UMEMOTO Rikuhei, NING Ding, FUJII Keisuke

Proceedings of the Annual Conference of JSAI Vol. JSAI2024 ( 0 ) page： 3Xin221 - 3Xin221 2024

Language：Japanese Publishing type：Research paper (scientific journal) Publisher：The Japanese Society for Artificial Intelligence

Ultimate is a sport where two teams of seven players compete to score points by passing a disc and catching it in the end zone. Unlike other sports, Ultimate has the unique characteristic that players holding the disc cannot move, emphasizing the importance of creating space to receive passes. While there is ample data and research on space in sports like soccer and basketball, such information is lacking for Ultimate. This study aims to focus on the widely adopted 3-vs-3 format in training by many teams and evaluate the space during offensive plays. The data used in this study was obtained through filming with a drone camera, and position data was acquired through angle correction and coordinate transformations. The model employed in this study is based on a soccer's pitch control model, which indicates the probability of controlling the ball at various points on the pitch. This model is adapted to suit the rules of Ultimate, where players cannot move while holding the disc. By combining weights that take into account position and distance with the values of the pitch control, space evaluation metrics are calculated, and the relationship between scoring and the space evaluation metric immediately before a pass was analyzed.

DOI： 10.11517/pjsai.jsai2024.0_3xin221

CiNii Research

Front line defensive evaluation based on event detection from full pitch video in soccer

IDE Kenjiro, UCHIDA Ikuma, UMEMOTO Rikuhei, FUJII Keisuke

Proceedings of the Annual Conference of JSAI Vol. JSAI2024 ( 0 ) page： 3Xin247 - 3Xin247 2024

Language：Japanese Publishing type：Research paper (scientific journal) Publisher：The Japanese Society for Artificial Intelligence

Most data analysis in soccer requires tracking data of players and ball and event data, and currently well-funded professional leagues are collecting a lot of data manually through companies. Previous research has used machine learning models to perform specific event detection from professional broadcast video, but amateur video is difficult to be analyzed due to the existence of various recording conditions. The purpose of this study is to perform specific event detection from amateur full-pitch video and to examine the validity of front-line defensive evaluation based on the detection results. In this study, two methods were used to detect events from video: the first is a deep learning-based e2e-spot method that uses video frames as input and predictive labels as output; the second is a rule-based method that uses ball velocity from ball tracking data and detects the start of a pass based on changes in velocity. We performed the fine-tuning of e2e-spot on amateur full-pitch video, was able to detect events more accurately than the rule-based method. We examined how accurately the estimated events can be used to evaluate front line defenses in soccer.

DOI： 10.11517/pjsai.jsai2024.0_3xin247

CiNii Research

Towards the Construction of a Soccer Foundation Model Using Simulation Data

SOMEYA Taiga, SCOTT Atom James, FUJII Keisuke, AKIYAMA Hidehisa, NAKASHIMA Tomoharu, YANAKA Hitomi

Proceedings of the Annual Conference of JSAI Vol. JSAI2024 ( 0 ) page： 3I5OS27b02 - 3I5OS27b02 2024

Language：Japanese Publishing type：Research paper (scientific journal) Publisher：The Japanese Society for Artificial Intelligence

Recent advancements in large foundation models have spanned various domains, including natural language processing, autonomous driving, and multivariate time series forecasting. Meanwhile, applied sports analysis has become widespread, with a particular focus on constructing quantitative evaluation methods for players and teams through the modeling of match situations. Despite this, a soccer foundation model capable of performing various tasks within a single model remains unexplored. This study explores a potential soccer foundation model by applying a multivariate time series prediction architecture for forecasting soccer trajectory data. We propose using log data from soccer simulation leagues for training, taking into account 1) the small scale of real trajectory data and 2) the effectiveness of synthetic data in constructing foundation models as indicated by previous research. Furthermore, we evaluate the effectiveness of the embedding representations by qualitatively comparing their similarities with actual soccer trajectories, confirming their applicability in downstream tasks.

DOI： 10.11517/pjsai.jsai2024.0_3i5os27b02

CiNii Research

　 PREV - NEXT 　

<FUJII Keisuke>

Papers - FUJII Keisuke