Large-Scale Multimodal Sports Data Analytics
Collection, management, integration, fusion, temporal modeling, data quality, domain adaptation, and scalable sports data pipelines.
Multimodal streams
Edge-cloud AI
Trustworthy decisions
2026 IEEE BigData Workshop
AI-Sports 2026
The Sixth International Workshop on Artificial Intelligence in Sports
A focused forum for large-scale sports data analytics, multimodal sensing, foundation models, edge AI, and trustworthy decision support across sports science and technology.
Nov. 7, 2026
Submission deadline
Nov 15, 2026
Acceptance notification
Nov 21, 2026
Camera-ready deadline
TBA
Workshop date
Phoenix, AZ, USA
Sheraton Phoenix Downtown
Overview
A data-centric forum for deployable sports intelligence
AI-Sports 2026 is proposed as a focused IEEE BigData workshop for research that treats sports intelligence as an end-to-end data problem. The workshop foregrounds the full lifecycle of sports data: acquisition from wearable, video, physiological, motion-capture, competition, and contextual sources; scalable data management; multimodal modeling; real-time inference; and accountable decision support for high-impact practice settings.
The workshop builds on five AI-Sports editions at IEEE ICME and complementary experience from IT4PSS at IJCAI. Its IEEE BigData positioning broadens the community beyond media analysis toward large-scale analytics, heterogeneous sensor integration, reproducible benchmarks, edge-cloud systems, foundation models, trustworthy AI, and deployment-ready sports science applications.
Volume
Large video, sensor, and event archives
Velocity
Real-time athlete monitoring and inference
Variety
Video, IMU, EMG, PPG, GPS, RFID, logs, and context
Veracity
Noisy, missing, drifting, and domain-shifted signals
Value
Actionable guidance for coaches, athletes, and clinicians
Call For Contributions
Original work, benchmarks, datasets, systems, and position papers
We invite research papers, position papers, benchmark papers, dataset papers, and system demonstrations. Submissions should emphasize data scale, multimodal integration, methodological novelty, reproducible evaluation, and practical relevance to sports science and sports technology.
Work addressing large-scale analytics, AI-driven data intelligence, IoT and edge data processing, trustworthy modeling, and deployable decision-support systems for real-world sports environments is especially welcome.
Topic Areas
Eight entry points into sports data intelligence
The scope connects sports applications with IEEE BigData themes in analytics, systems, AI, reproducibility, and real-world deployment.
Focus Area A
Large-Scale Multimodal Sports Data Analytics
Collection, management, integration, fusion, temporal modeling, data quality, domain adaptation, and scalable sports data pipelines.
Sports Video, Vision, and Spatiotemporal Understanding
Detection, tracking, pose estimation, action recognition, event detection, tactical analysis, motion forecasting, and long-horizon modeling.
Foundation Models, LLMs, and Agentic AI
Vision-language models, sports-aware foundation models, retrieval-augmented coaching, report generation, and human-in-the-loop workflows.
Big Data Systems, Edge AI, IoT, and Real-Time Computing
Wearable streams, edge-cloud coordination, latency-aware analytics, secure data transmission, and deployment constraints.
Trustworthy, Responsible, and Reproducible Sports AI
Calibration, robustness, privacy-preserving learning, fairness, transparent reporting, explainability, audit trails, and reproducible benchmarks.
Simulation, Digital Twins, and Biomechanics-Aware Analytics
Athlete, team, and environment digital twins; physics-aware learning; synthetic data; motion reconstruction; and what-if analysis.
Sports Applications and Human-Centered Systems
Performance analysis, training planning, rehabilitation assessment, return-to-play support, AR/VR feedback, and adaptive dashboards.
Explainable AI and Interpretable Sports Analytics
Feature attribution, visual explanations, counterfactual coaching guidance, interpretable multimodal fusion, and user-centered explanation interfaces.
Important Dates
Tentative 2026 timeline
IEEE BigData 2026 will take place Dec. 14-17, 2026 at Sheraton Phoenix Downtown in Phoenix, Arizona, USA. The exact workshop date will be announced after final IEEE BigData scheduling.
Submission countdown
-- days
Submission deadline: Nov. 7, 2026
Submission deadline
Notification of acceptance
Camera-ready deadline
Workshop date
Program
Preliminary one-day format
The program combines invited keynotes, oral papers, poster and demo exchange, and a cross-venue panel with academic and industrial perspectives.
Opening and overview
Community framing for sports data intelligence at IEEE BigData.
Opening and overview
Invited Keynote 1: Foundation models and agents in sports
Coffee break
Paper Session A: oral presentations, 15-18 minutes each
Lunch
Invited Keynote 2: IoT-based sports AI
Paper Session B
Poster and demo session: tools, datasets, benchmarks, live systems
Panel: From Perception to Agentic Sports AI
Closing remarks and community roadmap
Submission
Submit through the official conference system
All workshop papers must be submitted through the official submission system provided by the conference. Submissions sent directly to the workshop organizers will not be considered.
Submission Link
IEEE BigData 2026 CyberChair
Open Submission System
https://wi-lab.com/cyberchair/2026/bigdata26/scripts/submit.php?subarea=S24&undisplay_detail=1&wh=/cyberchair/2026/bigdata26/scripts/ws_submit.php
Organizing Committee
International organizing team
The team brings together AI-Sports, multimedia, sensing, machine learning, IoT, and sports science experience across Taiwan, the United States, Japan, and Germany.
Huang-Chia Shih
National Central University, Taiwan
shih@ncu.edu.tw Homepage
Tsì-Uí İk
National Yang Ming Chiao Tung University, Taiwan
tik@nycu.edu.tw Homepage
Min-Te Sun
National Central University, Taiwan
msun@csie.ncu.edu.tw Homepage
Wei-Shinn Ku
Auburn University, USA
weishinn@auburn.edu Homepage
Kazuya Sakai
Tokyo Metropolitan University, Japan
ksakai@tmu.ac.jp Homepage
Takahiro Ogawa
Hokkaido University, Japan
ogawa@lmd.ist.hokudai.ac.jp Homepage
Jenq-Neng Hwang
University of Washington, USA
hwang@uw.edu Homepage
Rainer Lienhart
University of Augsburg, Germany
rainer.lienhart@uni-a.de HomepageOrganizer profile highlights
Huang-Chia Shih works on multimedia content analysis, human-computer interaction, pattern recognition, medical image processing, and sports big data analytics. He is the lead workshop organizer.
Tsì-Uí İk focuses on intelligent sports learning, intelligent transportation systems, mobile sensing, machine learning, deep learning, and wireless sensor networks.
Min-Te Sun researches deep learning, network security, and distributed computing, and serves as chair of CSIE at National Central University.
Wei-Shinn Ku leads work in data management systems, data science, cybersecurity, and mobile computing as an endowed professor at Auburn University.
Takahiro Ogawa researches AI, IoT, and big data analysis for multimedia signal processing and applications.
Jenq-Neng Hwang is a fellow of IEEE and a long-standing contributor to multimedia signal processing, computer vision, AI, and industry-connected machine learning.
Rainer Lienhart leads the Machine Learning and Computer Vision Lab at the University of Augsburg, with expertise in large-scale video, human pose, sensor, and data mining algorithms.
Program Committee
Tentative PC members
Ming-Ching Chang, University at Albany, SUNY, USA
Jenq-Neng Hwang, University of Washington, USA
Chang-Tien Lu, Virginia Tech, USA
Wen-Chih Peng, National Yang Ming Chiao Tung University, Taiwan
Kazuya Sakai, Tokyo Metropolitan University, Japan
Huang-Chia Shih, National Central University, Taiwan
Tomotaka Wada, Kansai University, Japan
Jiunn-Lin Wu, National Chung Hsing University, Taiwan
Yu-Hsuan Kuo, Amazon, USA
Rainer Lienhart, University of Augsburg, Germany
Thomas B. Moeslund, Aalborg University, Denmark
Hideo Saito, Keio University, Japan
Anthony Cioppa, University of Liège, Belgium
Michele Merler, IBM Research, USA
Jim Little, University of British Columbia, Canada
Sho Takahashi, Hokkaido University, Japan
Shin'ichi Satoh, National Institute of Informatics, Japan
Chih-Chang Yu, Chung Yuan Christian University, Taiwan
Tiziana D'Orazio, National Research Council of Italy, Italy
Hsu-Yung Cheng, National Central University, Taiwan
Tzyy-Yuang Shian, National Taiwan Normal University, Taiwan
Sheng K Wu, National Taiwan University of Sport, Taiwan
History and Positioning
A bridge between sports AI communities and IEEE BigData
AI-Sports 2026 extends existing sports AI, computer vision, multimedia, and precision sports science communities toward large-scale, heterogeneous, real-time, and trustworthy sports data intelligence.
AI-Sports at IEEE ICME
The AI-Sports workshop series ran across five ICME editions, forming a community around sports analysis, multimedia understanding, sensing technologies, and intelligent sports applications.
IT4PSS at IJCAI
IT4PSS cultivated an IJCAI-side audience for intelligent technologies in precision sports science and complements AI-Sports' broader history.
CVsports and MMSports
CVsports and MMSports show sustained interest in sports vision and multimodal sports content. The IEEE BigData workshop complements them through systems, scalable analytics, IoT, reproducibility, and data-driven deployment.