Dagstuhl-Seminar 28051
Computational Modeling of Skeletal Data in Sports
( 30. Jan – 04. Feb, 2028 )
Permalink
Bitte benutzen Sie folgende Kurz-Url zum Verlinken dieser Seite:
https://www.dagstuhl.de/28051
Organisatoren
- Silvio Giancola (KAUST - Thuwal, SA)
- Matthias Kempe (Universität Wien, AT)
- Patrick Lucey (Stats Perform - Chicago, US)
- Maaike Van Roy (KU Leuven, BE)
Kontakt
- Marsha Kleinbauer (für wissenschaftliche Fragen)
- Simone Schilke (für administrative Fragen)
Advances in computer vision and pose estimation have transformed how athlete movement is captured, evolving tracking data from simple center-of-mass positions to detailed full-body skeletal models. This shift toward high-dimensional skeletal data creates unprecedented opportunities to study biomechanics, movement quality, tactical decision-making, and injury risk directly from how athletes move rather than merely where they move. However, it also raises new challenges around representation, evaluation, and translation into practice.
This Dagstuhl Seminar brings together researchers and practitioners from computer vision, machine learning, biomechanics, and sports science to explore how skeletal data can be represented, modeled, and used across the full pipeline from raw sensing to applied decision-making. Discussions will address: (1) effective representations and models for skeletal data, including graph-based, generative, and multimodal approaches; (2) how skeletal data can inform tactical, performance, and skill-related analyses; (3) its role in biomechanics, including load, joint mechanics, and injury-risk assessment; (4) evaluation of both raw skeletal data and downstream analysis tasks; (5) how findings can be communicated across disciplines and translated into practice; and (6) standards and regulatory guidance for validating skeletal data and derived insights for sporting federations and other stakeholders.
Building on two earlier Dagstuhl Seminars that established a shared vocabulary for sports analytics and explored event- and tracking-data-driven approaches to strategy and tactics, this seminar narrows its focus to the emerging modality of skeletal data. At the same time, it deliberately brings in a wider range of disciplinary backgrounds, including biomechanics, clinical movement science, and officiating, that have so far had limited engagement with the computational sports-analytics community.
The five-day program opens with high-level talks from academic, industry, and practitioner perspectives to establish common terminology across disciplines. The remainder of the week combines breakout sessions, deliberately composed of participants with complementary expertise in data generation, modeling, and applied domains, with focused technical presentations and panel discussions on translating computational findings into practice. A mid-week synthesis session will begin defining a cross-functional grand challenge spanning both the generation of skeletal data (robustness, calibration, uncertainty, transfer across sports and conditions) and its downstream use (injury-risk screening, skill assessment, officiating support, fatigue monitoring).
The seminar's central goal is to connect the communities that generate skeletal data with those that use it, and to agree on shared standards, interfaces, and evaluation criteria before these choices become fixed within separate research or commercial silos. Expected outcomes include a detailed specification for a public benchmark challenge, methodological recommendations, and a roadmap (supported by a steering committee and a shared communication channel) for continued collaboration between academia, industry, and sports practitioners in the 12–18 months following the seminar.
Creative Commons BY 4.0
Matthias Kempe, Maaike Van Roy, Patrick Lucey, and Silvio Giancola
Verwandte Seminare
- Dagstuhl-Seminar 06381: Computer Science in Sport (2006-09-17 - 2006-09-20) (Details)
- Dagstuhl-Seminar 08372: Computer Science in Sport - Mission and Methods (2008-09-07 - 2008-09-10) (Details)
- Dagstuhl-Seminar 11271: Computer Science in Sport - Special emphasis: Football (2011-07-03 - 2011-07-06) (Details)
- Dagstuhl-Seminar 13272: Computer Science in High Performance Sport - Applications and Implications for Professional Coaching (2013-06-30 - 2013-07-03) (Details)
- Dagstuhl-Seminar 15382: Modeling and Simulation of Sport Games, Sport Movements, and Adaptations to Training (2015-09-13 - 2015-09-16) (Details)
- Dagstuhl-Seminar 21411: Machine Learning in Sports (2021-10-10 - 2021-10-15) (Details)
- Dagstuhl-Seminar 24081: Computational Approaches to Strategy and Tactics in Sports (2024-02-18 - 2024-02-23) (Details)
Klassifikation
- Computer Vision and Pattern Recognition
- Machine Learning
Schlagworte
- Performance Analysis
- Biomechanics
- Sports
- Tracking Data
