This paper introduces a score-driven rating system, a generalization of the classical Elo rating system that employs the score, i.e. the gradient of the log-likelihood, as the updating mechanism for player and team ratings. The proposed framework extends beyond simple win/loss game outcomes and accommodates a wide range of game results, such as point differences, win/draw/loss outcomes, or complete rankings. Theoretical properties of the score are derived, showing that it has zero expected value, sums to zero across all players, and decreases with increasing value of a player's rating, thereby ensuring internal consistency and fairness. Furthermore, the score-driven rating system exhibits a reversion property, meaning that ratings tend to follow the underlying unobserved true skills over time. The proposed framework provides a theoretical rationale for existing dynamic models of sports performance and offers a systematic approach for constructing new ones.
Jointly forecasting trajectories of multiple interacting agents is a core challenge in sports analytics and other domains involving complex group dynamics. Accurate prediction enables realistic simulation and strategic understanding of gameplay evolution. Most existing models are evaluated solely on per-agent accuracy metrics (minADE, minFDE), which assess each agent independently on its best-of-k prediction. However these metrics overlook whether the model learns which predicted trajectories can jointly form a plausible multi-agent future. Many state-of-the-art models are designed and optimized primarily based on these metrics. As a result, they may underperform on joint predictions and also fail to generate coherent, interpretable multi-agent scenarios in team sports. We propose CausalTraj, a temporally causal, likelihood-based model that is built to generate jointly probable multi-agent trajectory forecasts. To better assess collective modeling capability, we emphasize joint metrics (minJADE, minJFDE) that measure joint accuracy across agents within the best generated scenario sample. Evaluated on the NBA SportVU, Basketball-U, and Football-U datasets, CausalTraj achieves competitive per-agent accuracy and the best recorded results on joint metrics, while yielding qualitatively coherent and realistic gameplay evolutions.
Sports video analysis is a key domain in computer vision, enabling detailed spatial understanding through multi-view correspondences. In this work, we introduce SoccerNet-v3D and ISSIA-3D, two enhanced and scalable datasets designed for 3D scene understanding in soccer broadcast analysis. These datasets extend SoccerNet-v3 and ISSIA by incorporating field-line-based camera calibration and multi-view synchronization, enabling 3D object localization through triangulation. We propose a monocular 3D ball localization task built upon the triangulation of ground-truth 2D ball annotations, along with several calibration and reprojection metrics to assess annotation quality on demand. Additionally, we present a single-image 3D ball localization method as a baseline, leveraging camera calibration and ball size priors to estimate the ball's position from a monocular viewpoint. To further refine 2D annotations, we introduce a bounding box optimization technique that ensures alignment with the 3D scene representation. Our proposed datasets establish new benchmarks for 3D soccer scene understanding, enhancing both spatial and temporal analysis in sports analytics. Finally, we provide code to facilitate access to our annotations and the generation pipelines for the datasets.
Motion instruction is a crucial task that helps athletes refine their technique by analyzing movements and providing corrective guidance. Although recent advances in multimodal models have improved motion understanding, generating precise and sport-specific instruction remains challenging due to the highly domain-specific nature of sports and the need for informative guidance. We propose CoachMe, a reference-based model that analyzes the differences between a learner's motion and a reference under temporal and physical aspects. This approach enables both domain-knowledge learning and the acquisition of a coach-like thinking process that identifies movement errors effectively and provides feedback to explain how to improve. In this paper, we illustrate how CoachMe adapts well to specific sports such as skating and boxing by learning from general movements and then leveraging limited data. Experiments show that CoachMe provides high-quality instructions instead of directions merely in the tone of a coach but without critical information. CoachMe outperforms GPT-4o by 31.6% in G-Eval on figure skating and by 58.3% on boxing. Analysis further confirms that it elaborates on errors and their corresponding improvement methods in the generated instructions. You can find CoachMe here: https://motionxperts.github.io/
Sarah Rebecca Ondraszek, Jörg Waitelonis, Katja Keller
et al.
An essential component for evaluating and comparing physical and cognitive capabilities between populations is the testing of various factors related to human performance. As a core part of sports science research, testing motor performance enables the analysis of the physical health of different demographic groups and makes them comparable. The Motor Research (MO|RE) data repository, developed at the Karlsruhe Institute of Technology, is an infrastructure for publishing and archiving research data in sports science, particularly in the field of motor performance research. In this paper, we present our vision for creating a knowledge graph from MO|RE data. With an ontology rooted in the Basic Formal Ontology, our approach centers on formally representing the interrelation of plan specifications, specific processes, and related measurements. Our goal is to transform how motor performance data are modeled and shared across studies, making it standardized and machine-understandable. The idea presented here is developed within the Leibniz Science Campus ``Digital Transformation of Research'' (DiTraRe).
Badminton, known for having the fastest ball speeds among all sports, presents significant challenges to the field of computer vision, including player identification, court line detection, shuttlecock trajectory tracking, and player stroke-type classification. In this paper, we introduce a novel video clipping strategy to extract frames of each player's racket swing in a badminton broadcast match. These clipped frames are then processed by three existing models: one for Human Pose Estimation to obtain human skeletal joints, another for shuttlecock trajectory tracking, and the other for court line detection to determine player positions on the court. Leveraging these data as inputs, we propose Badminton Stroke-type Transformer (BST) to classify player stroke-types in singles. To the best of our knowledge, experimental results demonstrate that our method outperforms the previous state-of-the-art on the largest publicly available badminton video dataset (ShuttleSet), another badminton dataset (BadmintonDB), and a tennis dataset (TenniSet). These results suggest that effectively leveraging ball trajectory is a promising direction for action recognition in racket sports.
Farhad Rezazadeh, Shirin Aali, Fariborz Imani
et al.
<i>Background and Objectives:</i> Chronic low back pain (CLBP) is associated with altered neuromuscular control. Dynamic Neuromuscular Stabilization (DNS) targets core–limb coordination; however, its specific impact on lower-limb electromyographic (EMG) activity during gait remains unclear. <i>Materials and Methods:</i> Fifty-five young adults with non-specific CLBP (pain ≥ 3 months with no identifiable specific pathology) completed the trial (overall mean age 23.7 ± 1.3 years). Participants were randomized to an 8-week DNS program or a control. Pre-/Post-intervention surface EMG during gait and clinical outcomes (VAS, ODI) were assessed. <i>Results:</i> Compared with control, DNS showed lower adjusted Post-test VAS (3.08 ± 0.25 vs. 6.13 ± 0.24; <i>ηp</i><sup>2</sup> = 0.596) and ODI (15.73 ± 1.55% vs. 34.36 ± 1.52%; <i>ηp</i><sup>2</sup> = 0.579). Directionally, DNS was associated with phase-specific EMG modulation: tibialis anterior during mid-stance was lower (<i>ηp</i><sup>2</sup> = 0.137), rectus femoris during push-off was lower (<i>ηp</i><sup>2</sup> = 0.119), biceps femoris during push-off was lower (<i>ηp</i><sup>2</sup> = 0.168), and vastus medialis at heel-strike was higher (<i>ηp</i><sup>2</sup> = 0.077) relative to control. Other muscle–phase pairs showed no adjusted between-group differences. <i>Conclusions:</i> An 8-week DNS program was associated with clinically meaningful reductions in pain and disability and with phase-specific changes in lower-limb EMG during gait. These findings support DNS as a promising rehabilitation option for young adults with CLBP; confirmation in larger trials with active comparators is warranted.
This case report describes the acceleration of an Olympic decathlete’s return to competition induced via high-frequency Blood Flow Restriction (BFR) training. BFR has gained popularity as an innovative rehabilitation method for promoting muscle repair and adaptation through anabolic and regenerative pathways when high mechanical loading is not possible. A 26-year-old elite decathlete with nine years of international experience sustained a Grade 2b strain of the semimembranosus and semitendinosus (a 9 mm central tendon tear) during a hurdle sprint. The injury was confirmed via MRI two days post-injury. Grade 2b hamstring injuries with intramuscular tendon involvement commonly require up to 4 weeks of rehabilitation before full training can be resumed. With the athlete due to complete in an Olympic Games competition 17 days post-injury, an intensive BFR-assisted rehabilitation program was initiated. Over 12 consecutive days, the athlete completed 3–6 BFR sessions per day (20–30 min each) at 50% limb occlusion pressure, along with physiotherapy and pain-limited functional testing. BFR was applied passively for recovery, during conditioning, and in low-load strength sessions. By day 12, sprint velocity reached 95% maximum, and the athlete successfully completed the decathlon, with no adverse effects or reinjury. This case illustrates how high-frequency BFR-assisted rehabilitation may facilitate accelerated recovery from a hamstring injury, enabling an effective return to elite competition within condensed timelines.
In team sports, effective tactical communication is crucial for success, particularly in the fast-paced and complex environment of outdoor athletics. This paper investigates the challenges faced in transmitting strategic plans to players and explores potential solutions using eXtended Reality (XR) technologies. We conducted a formative study involving interviews with 4 Division I professional soccer coaches, 4 professional players, 2 college club coaches, and 2 college club players, as well as a survey among 17 Division I players. The study identified key requirements for tactical communication tools, including the need for rapid communication, minimal disruption to game flow, reduced cognitive load, clear visualization for all players, and enhanced auditory clarity. Based on these insights, we propose a potential solution - a Mobile Augmented Reality (AR) system designed to address these challenges by providing real-time, intuitive tactical visualization and communication. The system aims to improve strategic planning and execution, thereby enhancing team performance and cohesion. This work represents a significant step towards integrating XR technologies into sports coaching, offering a modern and practical solution for real-time tactical communication.
Recent integration of Natural Language Processing (NLP) and multimodal models has advanced the field of sports analytics. This survey presents a comprehensive review of the datasets and applications driving these innovations post-2020. We overviewed and categorized datasets into three primary types: language-based, multimodal, and convertible datasets. Language-based and multimodal datasets are for tasks involving text or multimodality (e.g., text, video, audio), respectively. Convertible datasets, initially single-modal (video), can be enriched with additional annotations, such as explanations of actions and video descriptions, to become multimodal, offering future potential for richer and more diverse applications. Our study highlights the contributions of these datasets to various applications, from improving fan experiences to supporting tactical analysis and medical diagnostics. We also discuss the challenges and future directions in dataset development, emphasizing the need for diverse, high-quality data to support real-time processing and personalized user experiences. This survey provides a foundational resource for researchers and practitioners aiming to leverage NLP and multimodal models in sports, offering insights into current trends and future opportunities in the field.
Access to expert coaching is essential for developing technique in sports, yet economic barriers often place it out of reach for many enthusiasts. To bridge this gap, we introduce Poze, an innovative video processing framework that provides feedback on human motion, emulating the insights of a professional coach. Poze combines pose estimation with sequence comparison and is optimized to function effectively with minimal data. Poze surpasses state-of-the-art vision-language models in video question-answering frameworks, achieving 70% and 196% increase in accuracy over GPT4V and LLaVAv1.6 7b, respectively.
This study presents a novel contact detection system for "Parablade," a chambara-based, sword-play augmented sport. Augmented sports combine physical activities with virtual parameters (VPs) to create a balanced and equitable gaming experience, irrespective of players' physical capabilities. The proposed Parablade Microphone Unit (PMU) employs multiple micro-phones and machine learning algorithms to detect and classify hit events through sound recogni-tion. This system aims to ensure real-time updates of VPs, thereby enhancing the gameplay expe-rience. Experimental results indicate that the PMU can accurately recognize the occurrence and location of hit events with a high accuracy rate of 93.33%, with the assistance of 10kHz additional sound generated from the sword.
Video-based visual relation detection tasks, such as video scene graph generation, play important roles in fine-grained video understanding. However, current video visual relation detection datasets have two main limitations that hinder the progress of research in this area. First, they do not explore complex human-human interactions in multi-person scenarios. Second, the relation types of existing datasets have relatively low-level semantics and can be often recognized by appearance or simple prior information, without the need for detailed spatio-temporal context reasoning. Nevertheless, comprehending high-level interactions between humans is crucial for understanding complex multi-person videos, such as sports and surveillance videos. To address this issue, we propose a new video visual relation detection task: video human-human interaction detection, and build a dataset named SportsHHI for it. SportsHHI contains 34 high-level interaction classes from basketball and volleyball sports. 118,075 human bounding boxes and 50,649 interaction instances are annotated on 11,398 keyframes. To benchmark this, we propose a two-stage baseline method and conduct extensive experiments to reveal the key factors for a successful human-human interaction detector. We hope that SportsHHI can stimulate research on human interaction understanding in videos and promote the development of spatio-temporal context modeling techniques in video visual relation detection.
We develop a statistical framework to evaluate evidence of alleged cheating involving illegal signaling in sports from a forensic perspective. We explain why, instead of a frequentist procedure, a Bayesian approach is called for. We apply this framework to cases of alleged cheating in professional bridge and professional baseball. The diversity of these applications illustrates the generality of the method.
Satoshi Onoue, Noriaki Maeda, Yasunari Ikuta
et al.
Abstract The level of dynamic mechanical instability between the bony parts of the ankle joint provides important information on biomechanical function. However, the dynamics of the distance between the talus and lateral malleolus during gait remain unclear. This study aimed to compare the distance between the talus and lateral malleolus and the ankle joint angles during the stance phase of gait between individuals with chronic ankle instability (CAI) and healthy adults. The comparison was conducted using a synchronized ultrasound (US) imaging with a three-dimensional motion analysis (MA) system. This cross-sectional study included 12 participants (5 males, 7 females; age, 20.5 ± 1.8 years; height, 166.6 ± 9.4 cm; body weight, 60.2 ± 5.3 kg; body mass index, 21.7 ± 2.0 kg/m2; 16 feet) with CAI and 10 healthy controls (4 males, 6 females; age, 21.2 ± 1.6 years; height, 164.6 ± 10.5 cm; body weight, 56.8 ± 11.3 kg; body mass index, 20.8 ± 2.6 kg/m2; 20 feet). The distance between the talus and lateral malleolus during gait was significantly increased in the CAI group compared with that in the control group throughout the stance phase. The ankle dorsiflexion angle was smaller in the CAI group during the middle and terminal stance phases. Additionally, the ankle inversion angle was greater in the CAI group than in the control group. Our findings show the application of the synchronized US and MA system in the assessment of mechanical instability in CAI group, which may be used to determine treatment efficacy.