icSPORTS 2024 Abstracts

Area 1 - Computer Systems in Sports
Full Papers Short Papers
Area 2 - Health and Support Technology
Short Papers
Area 3 - Sport Performance and Support Technology
Full Papers Short Papers

Area 1 - Computer Systems in Sports

Full Papers
Paper Nr: 18
Title:

Evaluating Sensor Placement and Feature Importance for Hurling Movement Classification
Authors:

Chloe Leddy, Richard Bolger, Paul J. Byrne, Sharon Kinsella and Lilibeth Zambrano
Abstract: Human Activity Recognition (HAR) involves recognising and classifying human activities from data collected by sensors through machine learning (ML) techniques. The assessment of athletic movement via HAR has benefited sport performance analysis by identifying technical and tactical performance indicators. Hurling is a dynamic stick and ball invasion team sport that involves high-impact movements. Sensor placement and feature selection in HAR tasks impact the classification accuracy of the ML model during testing and training. This study aims to determine the optimal inertial measuring unit (IMU) sensor placement for recognizing hurling movements and to identify the most important features for accurate classification. Time-domain and frequency-domain features of accelerometer data were computed and were used to train and test three classification models: Support Vector Machine (SVM), Random Forest (RF) and k-Nearest Neighbour (k-NN). A RF model achieved the highest mean accuracy in the recognition of four hurling specific movements, for sensors located at the forearm (86%) and the thigh (84%). Features extracted from the z-axis, specifically zero crossing rate (ZCR), standard deviation (STD), and root mean square (RMS) were most discriminative in classifying hurling sport movements with a RF model using a forearm-mounted IMU.
Download
Paper Nr: 41
Title:

Component Replacement Study of 3D Human Pose Estimation Models in Real-World Complex Sports Scenarios: Focusing on Head Impact Events
Authors:

Yuchen Shi, Nobutake Ozeki, Ryu Yoshida, Motoki Inaji, Kazuyoshi Yagishita and Yusuke Miyazaki
Abstract: Human pose estimation in 3D is crucial in complex sports scenarios, particularly for athlete head impact events. We investigated the effect of different 2D pose estimation methods on the performance of 3D pose estimation models in complex sports environments. We used a transformer-based 3D human pose estimation model as a base framework, creating multiple variants by replacing the 2D pose estimator. These variants were evaluated using real sports game videos. Four 2D pose estimators were employed: Simple Baseline, High-Resolution Network (HRNet), Multi-stage Pose Network (MSPN), and Residual Steps Network (RSN). Performance was assessed using Mean per Joint Positional Error (MPJPE), Procrustes analysis MPJPE (P-MPJPE), and Mean per Joint Velocity Error (MPJVE) metrics. The results showed that MSPN performed the best in terms of position accuracy and motion velocity consistency (MPJPE, P-MPJPE and MPJVE). RSN presented promising absolute position accuracy (MPJPE) but showed limitations in the overall pose configuration (P-MPJPE). Simple Baseline and HRNet proved to be inadequate for complex sports scenarios. These findings indicate that different model architectures have different advantages in 3D human pose estimation in complex sports scenarios. This study provides insights for improving 3D pose estimation models in challenging real-world sports applications, contributing to the better understanding and prevention of sports-related head injuries.
Download
Paper Nr: 47
Title:

Analyzing Factors that Lead to NBA Regular Season Success
Authors:

Mohamad El-Hajj, Jackson Steed, Victor Gore, Craeg Jethro Infante, Raniel Flores, Danindu Wakista and Mohammed Elmorsy
Abstract: The National Basketball Association (NBA) values regular-season success and acknowledges the crucial role of a team’s roster composition in determining overall performance. This study uses machine learning techniques, specifically unsupervised learning clustering and decision tree models, to predict the composition of a winning roster. Our research identified three distinct clusters based on win percentage and the distribution of players across different skill levels. Successful teams typically have more top-tier players and a significant representation of players in the lowest skill level. In contrast, teams that spread their talent across the entire roster are less successful. We have noticed that players with average to above-average skills are notably affected by excessive playing time in the previous game, which leads to decreased performance and potential losses for the team in the next game. Considering the time of year and the gap between games, we recommend prioritizing the rest and recovery of top players, especially in the latter half of the season. It’s crucial to ensure that players who are not as skilled as the top players but still make significant contributions to the team maintain consistent performance, especially during the first half of the season. Analyzing height’s impact on basketball player performance has revealed practical insights that can empower coaches and management. We found that the shortest and tallest players often perform less than those of average height. Most top performers in the NBA tend to have heights closer to the average. However, for players who frequently operate near the net and encounter numerous rebound opportunities, it is generally preferable to have an average or taller player for slightly enhanced overall performance compared to below-average height players. Teams can use these insights to improve their roster construction and maximize player utilization by coaches from one game to the next. This research provides practical strategies that can be immediately implemented to enhance team performance.
Download
Paper Nr: 54
Title:

Assessing Physical Activity Levels While Playing Virtual Reality Exergames: A Pilot Study
Authors:

Mário Teixeira, Hildegardo Noronha, Pedro Campos and Cíntia França
Abstract: Due to the exponential growth in technology, exergames emerged as innovative tools that might be used to promote PA enjoyably. This study describes the development of a virtual reality (VR) exergame and the preliminary implementation results. The system was developed through the Unity3D platform and the HCT Vive, consisting of two mini games: a dance game and a snow skiing game. Five healthy adults (25.2  3.9 years) performed one VR exergame session and were monitored for PA intensity and heart rate (HR). After the session, participants were asked to report their perceived exertion and to fill in a system usability questionnaire. During the session, participants spent more time in sedentary activity (≈ 37.5%), followed by light activity (≈ 35.1%), and moderate-to-vigorous activity (≈ 27.4%). An average of 27.2 steps/min and HR of 123.5 bpm were registered while playing. Perceived exertion scores were higher in the dance mini game than in the snow skiing mini game. Regarding usability, participants considered the system easy to use and would like to use it more often. This study summarizes preliminary and promising results on the ability of VR exergames to promote light and moderate PA.
Download
Short Papers
Paper Nr: 11
Title:

Synthetic Data for Foot Strike Angle Estimation
Authors:

Christoph Schranz, Stefan Kranzinger and Stephanie R. Moore
Abstract: A runner’s foot strike angle (FSA) can be relied on to assess performance, comfort, and injury risk. However, the collection of FSA datasets is time-consuming and costly, which may result in small datasets in practice. Therefore, the creation of synthetic FSA datasets is of great interest to researchers to improve the performance of machine learning models while maintaining the same effort in data collection. We evaluate data augmentation (jittering, pattern mixing, SMOTE) and synthetic data generation (Generative Adversarial Networks, Variational Autoencoders) methods with four subsequent machine learning models to estimate the FSA on a dataset involving 30 runners across a range of FSAs. The results show promising results for the SVM and MLP, as well as for the jittering and pattern mixing augmentation methods. Our findings underscore the potential of data augmentation to improve FSA estimation accuracy.
Download
Paper Nr: 12
Title:

SportsNGEN: Sustained Generation of Realistic Multi-Player Sports Gameplay
Authors:

Lachlan Thorpe, Lewis Bawden, Karanjot Vendal, John Bronskill and Richard E. Turner
Abstract: We present a transformer decoder based sports simulation engine, SportsNGEN, trained on sports player and ball tracking sequences, that is capable of generating sustained gameplay and accurately mimicking the decision making of real players. By training on a large database of professional tennis tracking data, we demonstrate that simulations produced by SportsNGEN can be used to predict the outcomes of rallies, determine the best shot choices at any point, and evaluate counterfactual or what if scenarios to inform coaching decisions and elevate broadcast coverage. By combining the generated simulations with a shot classifier and logic to start and end rallies, the system is capable of simulating an entire tennis match. We evaluate SportsNGEN by comparing statistics of the simulations with those of real matches between the same players. We show that the model output sampling parameters are crucial to simulation realism and that SportsNGEN is probabilistically well-calibrated to real data. In addition, a generic version of SportsNGEN can be customized to a specific player by fine-tuning on the subset of match data that includes that player. Finally, we show qualitative results indicating the same approach works for football.
Download
Paper Nr: 17
Title:

Design and Implementation of a Stone Rotation Measurement System with IMU Sensor and Stone Behavior Presentation System
Authors:

Chen Guanyu, Haruna Mori, Yoshinari Takegawa, Shimpei Aihara and Fumito Masui
Abstract: In the strategic sport of curling, referred to as “chess on ice,” precise measurement and real-time presentation of stone dynamics are critical for performance optimization. This study introduces a novel system that integrates Inertial Measurement Units (IMUs) and real-time data processing to track and analyze the rotational dynamics of curling stones accurately. Additionally, our system includes the Stone Behavior Presentation System, which interactively displays real-time data on a tablet device, providing instant feedback to players and coaches. This dual approach not only enhances the accuracy of data collection but also improves the immediacy and applicability of the information for strategic decision-making in training and competitive environments.
Download
Paper Nr: 24
Title:

Creation of Training Data and Training for Prediction Model of Curling Scores Using Real Game Data
Authors:

Tomoya Iwasaki, Wataru Noguchi, Yasumasa Tamura, Shimpei Aihara and Masahito Yamamoto
Abstract: Curling is a sport in which two teams take turns shoting stones at each other on an ice field to compete for total scores. Curling is a highly strategic sport, and the strategy of stone delivering has a significant impact on the outcome of the game. To verify strategy of curling, “digital curling” is a platform that reproduces curling on a computer. Following the previous research of curling AI using game tree search and evaluation function by Ataka et al., real game data was obtained and trained into a neural network of evaluation function. In this study, we propose a method to obtain stone position information from real game data. Also, the model was trained from the obtained data. The results show that models trained with realistic data correspond better to realistic situations than conventional models trained with data generated by algorithms. However, in situations where there are many stones on the sheet, the model was also found to be insufficiently accurate as is the case with conventional models.
Download
Paper Nr: 43
Title:

Punch Type Classification and Hit Judgement Using Estimated Skeletal Model in Boxing Match Videos
Authors:

Soma Watanabe and Yoshinari Kameda
Abstract: In boxing, the choice of punch types and how to hit the punch to the opponent player is an important issue. So the support of computer vision on punch type classification and hit judgement on boxing match video is demanded. There are currently two challenges to that purpose. The first is the preparation of appropriate video dataset of boxing matches. The second is the discussion of the right method for punch type classification and punch hit judgment. We propose to create a video dataset of boxing matches from a boxing 3DCG simulation. The simulation can automatically annotate attributes to the dataset. This is useful for hit / no-hit judgement as it is not easy to identify which punches are actually hit and which are not on real boxing match videos. Based on the dataset we prepared, we propose a new method using time-series skeletal representation for classifying the type of punches and judging the hits. The experimental results show that our proposed method is able to classify the types of punches and judge the hits.
Download
Paper Nr: 44
Title:

Quantification of Visual Search Motion During Basketball in VR Simulation
Authors:

Shinya Ishikawa, Hidehiko Shishido, Kenji Yoshida and Yoshinari Kameda
Abstract: In basketball, players must be able to make good decisions in complex and rapidly changing situations. An important element of situational judgment is visual search motion. We propose to quantify the visual search motion of a basketball player during play. In the proposed method, the experience of playing basketball is realized on a VR simulation. The player wears a head-mounted display to experience the playing situation. We prepared scenarios that considered the characteristics of the basketball game and the characteristics of the head-mounted display and measured the visual search motion on the prototype system. Experiments were conducted on defensive players, and the results show that the measurement results of visual search motion are useful for discussing the characteristics of players.
Download
Paper Nr: 56
Title:

Leveraging AI to Mitigate Risks in Yoga Practice: A Real-Time Posture Correction Application
Authors:

Kushal Ganesh and Amar Ramudhin
Abstract: The extensive propagation of yoga, once reserved for isolated spiritual communities, has brought several advantages as more and more people have gained access to a holistic well-being system. Nonetheless, while many individuals have benefited from the influence of yoga, lax application of its various regimens has resulted in a hike in yoga-related injuries. The following paper explores current AI tools for yoga and then elucidates the physical, psychological, and long-term adverse effects of improper yoga practice, with supporting data and statistics. Furthermore, it suggests an AI-equipped application that alters the potential of injuries arising from incorrect body movements during yoga. Next, particular technical features of the app are enumerated, demonstrating how machine learning aids in data analytics to prevent misapplication and take care of user privacy issues. Finally, the challenges of the proposed solution's adoption, usability, and ethical usage are addressed, and suggestions are offered to circumvent such hurdles.
Download
Paper Nr: 65
Title:

Analysis of the 'Zero Step' IHF Handball Rule 7.3.d. in Open and Closed Motor Performance Program
Authors:

Igor Gruić and Ivan Vlah
Abstract: Handball is a dynamic team sport game in which a series of actions can take place in a short period of time and in a small space, without the human eye being able to follow it. In this paper, the 'zero step' in open (matches) and closed (training) motor performance program was analysed and an attempt was made to prove that it is not possible to simultaneously land on two feet, which significantly changes the rules of the handball game. The matches of the European Handball Championship 2024 held in Germany were analysed in the matches that ended in a draw, and whether referee errors related to unjudged steps during the performance of the 'zero step' could have affected the final result. 18 senior players of the Kozala handball club, aged 16 to 24, participated in the research. For the purposes of the research, a motor task was composed in which, in the first phase, there was a jump to the 'zero step' from a standing position and a run, and in the second phase, a given feint and a shot at the goal. The results showed that no player jumped on both feet at the same time. A qualitative analysis of the matches of the European Championship in 2024 came to the conclusion that the final outcome could have been different, if there were no refereeing errors during the performance of the zero step. According to the obtained results, the anthropometric characteristics do not affect the difference in the time between the contacts of the left and right foot with the ground during the performance of the closed structure of the 'zero step' from the initial stance and after the run-up.
Download
Paper Nr: 59
Title:

Estimation of Overlapped Tactical Actions from Soccer Match Video
Authors:

Kento Kuroda, Ikuma Uchida, Keisuke Fujii and Yoshinari Kameda
Abstract: In a soccer match, there may be overlaps of tactical actions performed at a certain point in time, because two teams take different tactical actions and each tactical action has no clear temporal boundary. In this study, we propose a new method for the estimation of overlapped tactical actions from soccer match videos. We enable the estimation of overlapped tactical actions considering exclusive relationships. We achieve this by having the deep learning model learn all tactical actions simultaneously. We validate this method using data from 10 matches. We succeeded in expressing the tactical actions performed at a given time in terms of the strength of several tactical actions.
Download

Area 2 - Health and Support Technology

Short Papers
Paper Nr: 21
Title:

Healing the Everyday Athlete, Evaluating the Impact of Recovery Wear Garment Use on Musculoskeletal Health in an Occupational Setting
Authors:

Kara Radzak and John Mercer
Abstract: While musculoskeletal injuries are commonly associated with sports, work-related musculoskeletal disorders greatly impact the workforce efficacy and account for a high volume of workers compensation cases. This study evaluated the effect of a carbonized charcoal and germanium recovery wear product on reducing musculoskeletal discomfort among first responders, a group prone to overuse injuries. Participants included 13 firefighters and law enforcement officers who used the recovery wear product over two months. Measures of musculoskeletal function and discomfort were assessed using the Fusionetics® Movement Health Questionnaire. Results indicated no change in measures of musculoskeletal function and discomfort over the 2-month period. Despite a small sample size of the present study, it is important to research the potential of recovery wear products as a practical intervention for managing work-related musculoskeletal disorders in physically demanding professions. Further research with larger samples and objective measures is needed to confirm these preliminary results and explore long-term benefits. This study provides a foundation for future investigations into wearable technology for occupational health.
Download
Paper Nr: 58
Title:

The Effects of Exergaming on Executive Functions in Children with ADHD: A Protocol of Systematic Review and Meta-Analysis
Authors:

Febryansah Gilang Aris Pradana, Rakhmat Ari Wibowo and Graham Baker
Abstract: Background: Attention-deficit/hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders that affect children. Symptoms of ADHD have been associated with reduced executive functions (EFs). Physical activity (PA) was suggested as an optional treatment; however, adherence to PA in children with ADHD is low. Exergaming has shown the beneficial effects on EFs; however, the evidence-base results were varied. This study will review primary studies investigating the effects of exergaming to increase EFs. Methods: This protocol of systematic review and meta-analysis is developed following the PRISMA guidelines. The study protocol has been registered in PROSPERO (CRD42024549395). Studies from eleven databases, published articles from the same authors, forward and backward, and co-citing and co-cited citation searches will be screened by two researchers based on inclusion and exclusion criteria. Two researchers will assess the risk of bias using The Cochrane Risk of Bias 2. Narrative synthesis will be presented, and meta-analyses will be performed on EF outcomes using random-effect meta-analyses. GRADE will be used to assess the quality of evidence by two researchers. Conclusion: This study will be the first systematic review and meta-analysis which offer insights on the effects of exergaming on EFs and adherence to PA in children with ADHD.
Download
Paper Nr: 60
Title:

Characterizing Locomotor Activity and Internal Load in VR-Based Exergames for Post-Musculoskeletal Injury Rehabilitation
Authors:

Élvio Rúbio Gouveia, Pedro Campos, Krzysztof Przednowek, Andreas Ihle, Adilson Marques, Hugo Sarmento, Diogo Martinho and Bruna Gouveia
Abstract: This study investigated the use of VR-based exergames in musculoskeletal injury rehabilitation, focusing on locomotor activity and internal load, as well as the influence of involvement, sensory fidelity, and interface quality. Thirty-seven participants (aged 19-53) engaged in five customized VR games designed for rehabilitation. These games included ”Weight Transfer,” ”Military March,” ”Side Squat,” ”Progressive March,” and ”Walking along a Straight Line.” Data were collected using HTC Vive Pro hardware and full-body tracking, with the E4 wristband measuring heart rate and movement and the OMNI scale assessing perceived exertion. The Witmer-Singer Presence Questionnaire evaluated user experience. The results revealed significant differences in heart rate, movement, perceived exertion, and exercise intensity across the exergames. ”Progressive March” produced the highest heart rate and intensity, while ”Side Squat” generated the most movement and exertion. ”Weight Transfer” had the lowest values across all metrics. Additionally, higher levels of presence in the VR environment were linked to more physical activity. These findings suggest that this VR-based exergame session effectively meets each rehabilitation phase’s needs. Higher Presence in VR enhances user engagement and realism, leading to increased physical activity.
Download
Paper Nr: 35
Title:

The Stress Is Real: Physiological Measurement of League of Legends Players Experience During a Live Esports Event
Authors:

David Berga, Eleonora De Filippi, Arijit Nandi, Eulàlia Febrer Coll, Marta Reverté Bel, Lautaro J. Russo and Alexandre Pereda-Baños
Abstract: Videogames and Esports experienced a huge growth in popularity lately and have opened a ripe new field for the study of human behavior. Esports gaming is an area in which videogame players need to cooperate and compete with each other, influencing their cognitive load, processing, stress, and as well as social skills. In this observational study we inquire whether variations in autonomic nervous system activity can be obtained reliably during a live League of Legends (LoL) event, especially considering this is a preliminary study with a limited participant sample. We found that game performance (winning or losing the game) significantly affects electrodermal activity and cardiac modulation, where players who lost the game showed higher stress-related physiological responses, compared to players who won. We also found that specific important events in the game, such as "Killing," "Dying," or "Destroying the turret," increased players' electrodermal and cardiac modulation compared with other less relevant events, such as "Placing the guards" or "Destroying the turret plates." Finally, by analyzing activity according to players' roles, we found various notable activity trends. Altogether, these (yet preliminary) results encourage further exploration of physiology-based applications for LoL and Esports players on live events.
Download
Paper Nr: 42
Title:

Sensor Module Application in the Design of an Automatic Tabletop Hockey Product Prototype for Leisure Activities and Mental Fitness
Authors:

Li-Shiue Gau, Chung-Wen Hung, Chia-Jui Yang, Tzu-Ling Liu and Shiou-Ling Lin
Abstract: With the fast-paced modern lifestyle and increasing work pressure, the demand for leisure and entertainment is rising. Traditional tabletop hockey has been a popular leisure activity, primarily providing relaxation and enjoyment while promoting interpersonal interaction. However, modern needs have evolved, transforming games that once required multiple participants into formats suitable for solo play. This evolution has led to the emergence of the automatic tabletop hockey product, creating new business opportunities. Despite its immense potential, there are currently no automatic tabletop hockey products available in the market. Therefore, the purpose of this study is threefold. First, sensor module applications were used to design an automatic tabletop hockey product prototype. Second, this study also conducts an in-depth practical investigation to assess the commercial feasibility of automatic tabletop hockey product in the perspective of leisure activities. Third, an experimental design was conducted with young elementary school children to see the experiential value of this product prototype, referring to the mental fitness. The anticipated contribution of this study not only aims to satisfy the modern demand for leisure and entertainment but also has the potential to open new business prospects.
Download

Area 3 - Sport Performance and Support Technology

Full Papers
Paper Nr: 13
Title:

Monitoring and Control of Power Preparedness of Athletes in Flatwater Rowing and Canoeing Using Strain Gauges
Authors:

Eugene Alooeff, Daniil Guseinov, Dmitry Lukashevich, Dmitry Bykov and Alexander Minchenya
Abstract: Control over the preparedness of flatwater rowers and canoeists is realized in the process of solving a variety of particular problems related to the organization of training activities, planning and dosing of loads, selection of training tools and methods for assessing various aspects of readiness and competent interpretation of the results obtained for carrying out corrective measures. Solving these problems is facilitated by strain gauges, which make it possible to record dynamic and some kinetic parameters in natural rowing conditions (on water recording) as part of training and monitoring activities. The article presents the developed strain gauge sensors, describes the features of their calibration and mounting on an athlete’s paddle, and also proposes software for automated processing of recorded data. The article is based on a practice-oriented study on experimental testing of the developed sensors as part of the training and monitoring process.
Download
Paper Nr: 15
Title:

Development of a Simple Tracking System to Monitor Curling Stone Dynamics
Authors:

Shimpei Aihara, Guanyu Chen, Riku Hara, Minori Nakagawa, Ayumi Ogasawara, Hitoshi Yanagi, Yoshinari Takegawa, Takeshi Ito and Fumito Masui
Abstract: Curling, frequently called chess on ice, is a sport renowned for its strategic depth, making it essential to analyze stone dynamics for advanced tactical purposes. Systems have been developed to meticulously measure stone position and dynamics using camera imagery and devices affixed to the rocks. However, systems relying on multiple dedicated cameras and mounted devices encounter challenges related to portability and simplicity. Therefore, this paper presents the development of a simple tracking system employing a single camera to measure stone position and dynamics. The system uses deep learning for stone detection and tracking and calibration for coordinate calculation to visualize stone dynamics. This paper aims to support training and tactical planning using this tracking system. Experiments were conducted using the tracking system to assess its accuracy for static stones and the velocity of moving stones. The results indicated an average positional accuracy error of 0.02 m and an average velocity accuracy of 0.05 m/s, demonstrating the tracking system’s high accuracy and practical feasibility.
Download
Paper Nr: 19
Title:

A Player Position Tracking Method Based on a Wide-Area Pan-Tilt-Zoom Video
Authors:

Shunzo Yamagishi, Chun Xie, Hidehiko Shishido and Itaru Kitahara
Abstract: This paper proposes a method to estimate the posture of an athlete moving on a vast field in a sporting event using a pan-tilt-zoom camera. In order to estimate the posture of an athlete on a sports field from a dynamic video sequence, our method extracts image features to search correspondence points among successive frames and computes the homography transformation matrix that compensates for changes the camera parameters (e.g., angle of view and posture). The effectiveness of this method is qualitatively verified using MLB TV broadcast video. The accuracy and error factors are also quantitatively verified by CG simulation.
Download
Paper Nr: 22
Title:

Performance Analysis in Volleyball: Problem of Merging Attack and Counterattack Spike as One Variable
Authors:

Ivana Klaričić, Hrvoje Ajman and Josip Cvenić
Abstract: The purpose of this study was to present the multiple regression model with attack and counterattack spike as separate variables is more appropriate than the model with the attack and counterattack spike merged as one variable. Two multiple regression analyses were conducted that determine the relationship between situational efficiency parameters of volleyball game phases with the set score. Game phases included into both regression models were the serve, reception, spike, block and dig. One of the regression models had an attack and counterattack spike as two separate variables and the other one had them merged as one. A sample was 40 randomly selected volleyball sets played in the European League for Men in 2011 and 2012. Although the sample wasn't recent, the purpose of this methodologically based study was to present deficiency of merging attack and counterattack spike as one variable. Both multiple regression analyses determined a high and positive relationship between the situational efficiency of volleyball game phases with the set score. The spike as a merged variable had 32.5% of common variance with the set score. But when separated into two variables, the attack spike had only 8.9% of common variance and counterattack spike 26.5%. Although the spike was the game phase that had the highest relationship with the set score, the spike in the counterattack was the one that contributed. Ultimately the serve, reception and dig had higher common variance with the set score then the attack spike. Conclusion was that the attack and counterattack spike need to be considered as separate variables because of specificity of the attack and counterattack complexes of the volleyball game.
Download
Paper Nr: 38
Title:

autoWT: A Semi-Automated ML-Based Movement Tracking System for Performance Tracking and Analysis in Olympic Weightlifting
Authors:

Gatis Jurkāns, Anikó Ekárt and Ulysses Bernardet
Abstract: As part of AI Coaching Assistant project research in sports performance tracking systems, we present autoWT, a novel semi-automated computer vision tracking system designed and developed for repeated long-term performance tracking of Olympic Weightlifting (OW) training. The system integrates multiple cameras and a heart rate sensor to capture, detect, and analyse OW movements, providing coaches and athletes with objective performance metrics. Key features include automated lift detection, clip extraction, and acquired performance metric visualisation based on markerless pose estimation data. The system architecture, consisting of a distributed system with multiple workers and a controller, enables efficient processing of high-bandwidth data streams. The paper provides an overall system architecture, operating principles and a detailed breakdown of action onset recognition and performance metric extraction system modules. We evaluate the system’s lift detection accuracy and the repeatability of extracted performance metrics using data from Olympic lifts. Results demonstrate high accuracy in lift detection and consistent and repeatable metric extraction, indicating autoWT’s potential as a valuable tool for conducting long-term Olympic weightlifting performance analysis studies and as an aid for coaching. The autoWT system can enhance the broader perspective and be an exemplary model for designing tracking systems in other sports.
Download
Short Papers
Paper Nr: 7
Title:

Optimizing Youth Basketball Training Through the Novel and Entertaining Approach of Self-Perception Evaluation Channel
Authors:

Mengru Liu, Anthony Kong, Newman Lau, Refati Rehe, Zeping Feng, Xi Liu and Kun-Pyo Lee
Abstract: In basketball training, players are required to make quick and accurate decisions while maintaining clear self-perception of competence in the game. This study proposed the basketball perception evaluation channel as an innovative approach to adjust players’ perceptions of competence. Twelve youth male players participated in the study, which involved video feedback and questioning tests, to assess their decision-making awareness and self-confidence level in offensive, defensive and overall performance. The results indicated that there were differences between the self-confidence levels and decision-making awareness of players, revealing instances of overconfidence or underconfidence in their basketball abilities and highlighting inaccuracies in the players' perceptions of competence. The basketball self-perception evaluation channel demonstrated a fun and creative approach for evaluating and observing the abilities and psychological states of players, aiding them in becoming aware of their actual basketball abilities and confidence states, and assisting coaches in providing targeted training and team management.
Download
Paper Nr: 20
Title:

A Human Pose Estimation Method from Pseudo-Captured Two-Viewpoints Video
Authors:

Ayaka Shimazu, Chun Xie, Satoru Tanigawa and Itaru Kitahara
Abstract: Motion analysis utilizing human pose estimation has garnered increasing attention within sports science, serving both preventive medicine and skill enhancement purposes. While techniques using 3D trackers and RGB-D cameras to estimate human poses are gaining attention, the widespread adoption is hindered by the requirement for extensive space and specialized equipment. This paper introduces a novel method to estimate the 3D human pose using RGB video data captured from 'pseudo' two-viewpoints. This approach involves performing the same motion in different directions and recording with a single camera. We confirm that the accuracy of 3D human pose estimation from video taken by a single camera is improved by the pseudo-two-viewpoints recording compared to existing methods using a single monocular RGB camera.
Download
Paper Nr: 23
Title:

The Development of Automatic Training Analysis Using 3D Accelerometer in Male Young Elite Soccer Team
Authors:

Takuya Magome, Takeshi Tanaka and Toru Yamaguchi
Abstract: In recent years, there has been a growing need to introduce IT technology to improve performance in professional sports and athlete training. However, while technological innovation is taking place, the amount of data that can actually be acquired is becoming enormous, and unless there are people who can provide expert advice, the compatibility with training is not high. In this study, we aimed to build a system that uses a three-dimensional accelerometer to appropriately evaluate the performance of athletes during training. We developed a classification function that detects and separates intervals from the measured time series data and automatically divides the separated sections by type of training. As a result of evaluation using data from 16 days of training for a U-18 soccer team (n=33), intervals were detected with 100% accuracy, and five types of training were detected with 80% accuracy. We confirmed that a system equipped with the developed function could speed up feedback to athletes.
Download
Paper Nr: 26
Title:

5G-Based Body Sensor Network for Real-Time Feedback in Running
Authors:

Sebastian Mayr and Harald Rieser
Abstract: Immediate feedback is vital in sports and motor learning. Traditionally, coaches provide this feedback. However, recent developments in wearable sensing and feedback devices, state-of-the-art processing algorithms and sensor data integration allow for new applications and insights. While this allows for more direct feedback, it also introduces new demands on feedback systems in terms of network transmission speed, energy consumption and needed calculation power. The 5G networking standard is a promising technology to address these issues. It promises reduced energy usage, high availability, latency in the low millisecond range and high data rates up to 10 Gbps. Despite these characteristics, there has not been much research on the use of 5G in sports biofeedback systems. Therefore, this work aims to assess the usability of 5G for a real-time feedback system in the sports context. We did this by developing a distributed feedback system for the assessment of running gait symmetry. The system utilizes internal smartphone IMU as sensors and employs a 5G infrastructure for data transmission to an offsite server for the calculation of running metrics and generation of the feedback signal. Pilot tests, using the feedback system, showed mean roundtrip times of 140.14ms (sd = 14.47ms) for outdoor use and 128.92ms (sd = 25.83ms) for indoor use. These results indicate that promised low latencies by the 5G standard are currently not reachable outside of controlled testing environments. Nonetheless, as these times are below average reaction times in recreational athletes (150ms) it is still promising for real-time feedback applications in running.
Download
Paper Nr: 28
Title:

Real-Time Tactical Analysis: Leveraging GNSS Position Data for Tactical Behavior
Authors:

Paolo Roberto Gabrielli and Stefano D'Ottavio
Abstract: In team sports such as football, tactical analysis has become essential to evaluate the collective behavior of the team and provide immediate and reliable feedback to the coaching staff. Global navigation satellite system (GNSS) tracking systems are widely used in team sports such as football. From geographic coordinates (e.g., latitude and longitude), information about the positions of players during a match can be obtained. Most GNSS-based systems use location tracking information to evaluate physical performance in terms of external load and physical stress. This research aims to present a technological innovation called GLF (Geo Live Football), which integrates tactical measures from raw GNSS data to understand the behavior of the team during matches, introducing the concept of "live" analysis.
Download
Paper Nr: 40
Title:

Towards a Smart Luge that Measures Steering Input of the Rider
Authors:

Rupert Staudinger, Wolfgang Kremser, Christoph Thorwartl and Hermann Schwameder
Abstract: Luge is a high-speed Olympic Winter sliding sport that is timed in milliseconds. The athlete’s steering performance is a crucial factor for success, but there are currently no objective methods to evaluate steering technique and timing. As a work in progress, we present a lab prototype of the ‘smart luge’, a sled retrofitted with six unobtrusive commodity force sensors. The results of a laboratory test with five simulated runs show that the current setup is capable of measuring the athlete’s activity during steering. This work aims to advance data-supported training in the luge sport by enabling the in situ measurement of luge athletes’ activity.
Download
Paper Nr: 52
Title:

Enhancing Speed Climbing Performance and Optimizing Training Methods Through Advanced Video Analysis
Authors:

Dominik Pandurevic, Pawel Draga, Klaus Hochradel, Lewis Chew, Muhammad Hidayat and Alexander Sutor
Abstract: This paper presents the implementation of an automated recording and analysis system for the capturing and evaluation of performances of speed climbing athletes. In collaboration with the National Youth Sports Institute (NYSI) in Singapore, an advanced camera system was installed on a newly constructed speed climbing wall, aiming to enhance training protocols by providing detailed performance insights. The paper is mainly split into two parts: the first one describes the hardware setup, including camera selection, configuration, integration with the existing infrastructure and data collection methods. The second part presents the analysis of performance data of one athlete trained at the NYSI, highlighting key findings and potential training improve-ments. Preliminary results indicate significant benefits in technique refinement and performance optimization, demonstrating the system’s value in a competitive training environment. Through section wise analysis of several runs of the same athlete, a stagnation in the start section of the standardised wall has been detected. However, a significant improvement was determined in the other sections, which led to a noticeable overall increase in performance in less than 2 years of regular training.
Download
Paper Nr: 61
Title:

Data Visualization for Dynamic Strength Index: A Qualitative Approach for Enhanced Interpretation and Decision-Making
Authors:

Zane Šmite, Artūrs Paiķis and Līga Plakane
Abstract: Dynamic strength index (DSI) serves as an important metric for assessing the balance between athletes maximal and ballistic strength. However, its interpretation can be limited if the effects of both isometric and ballistic components are not considered. The aim of this study is to develop a qualitative approach for interpreting DSI through data visualization, providing strength and conditioning coaches’ clearer insights that may guide more effective training recommendations. Thirty male college-level basketball athletes performed countermovement jumps (CMJ) and isometric mid-thigh pulls (IMTP) as a part of late-season testing. The peak IMTP force normalized to body mass was 39.02±3.57 N/kg, while peak CMJ force was 25.57±2.92 N/kg. The mean DSI was 0.66±0.09, where 2 athletes attained a DSI ≥0.80, 19 athletes between 0.60 and 0.80, and 9 athletes ≤0.60, corresponding to recommendations for maximal strength, concurrent, and ballistic training, respectively. T-score adjustments, used to categorize athletes based on maximal strength, resulted in the reclassification of 5 athletes from the concurrent training group to the maximal strength development group, and 5 athletes from the ballistic training group to the concurrent training group. Visualizing the DSI in a scatter plot, alongside T-score performance bands from the IMTP, allows for better evaluation of athletes' weaknesses and may guide more effective strength training recommendations.
Download
Paper Nr: 63
Title:

Y–Balance Test in Female Gymnasts
Authors:

Lucija Milčić, Elena Milenković and Josipa Radaš
Abstract: The aim of this study was to examine the differences in dynamic balance between the legs and potential asymmetry in rhythmic gymnasts. The sample consisted of six rhythmic gymnasts competing at the senior level A category and members of the Croatian National Team. Dynamic balance was assessed using the "Y Balance" test. Each gymnast performed the test in the anterior, posteromedial, and posterolateral directions, following a practice attempt. Data analysis was conducted using Statistica 14. Differences in Y Balance test performance between the dominant and non-dominant leg were evaluated using the T-test for dependent samples, as well as the Wilcoxon Matched Pairs Test for variables that did not follow a normal distribution. The results revealed a statistically significant difference in the posteromedial reach distance between the dominant and non-dominant legs, indicated by variables RRDPMD and RRDPMND (p=0.021). This disparity was evident in both absolute and relative terms, with the dominant leg demonstrating a greater reach distance. These findings suggest that the gymnasts exhibit a preference for one leg over the other, potentially affecting performance and increasing the risk of injury. Monitoring these asymmetries is crucial for developing targeted training interventions to enhance balance and functional performance.
Download
Paper Nr: 14
Title:

Biomechanics of the Lower Extremity in Youth Football League: FIFA 11+ One Leg Squat Analysis
Authors:

Anna Davidovica, Sergejs Davidovics, Guna Semjonova, Alexei Katashev, Alexander Oks, Linda Lancere, Signe Tomsone and Maksims Zolovs
Abstract: Background: Football carries substantial injury risks, especially for youth players. Providing biofeedback of lower limb motion during functional tasks is a crucial part of injury prevention programs such as FIFA 11+. While the FIFA 11+ warm-up program providing individualised feedback remains challenging, wireless sensor systems such as the DAid® Pressure Sock system, NOTCH® Inertial Sensor System, and PLUX Wireless Biosignals (muscleBAN kit) System offer potential solutions. Aim: This study aims to explore the correlation of lower limb biomechanical variables during the FIFA 11+ Part 2 exercise "One Leg Squat" in youth football players using wireless sensor systems and video recordings. Methods: Using wireless sensor systems and video recordings, we analysed lower limb biomechanics during the "One Leg Squat" exercise in youth football players. Results: Our findings showed significant associations: hip adduction correlated positively with foot plantar pressure centre (COP1y), internal hip rotation correlated strongly with COP1y and COP2x, and gluteus medius activity. Conversely, COP1y showed a negative correlation with gluteus maximus activity. Conclusions: The results support the potential of wireless sensor systems in monitoring the biomechanical changes of the lower extremity movements and lay the groundwork for future biofeedback methods based on the DAid® smart socks system technology for evaluating lower limb motion, especially the changes in the centre of pressure of the plantar surface of the foot, during functional tasks in Football Youth League players.
Download
Paper Nr: 27
Title:

Motion Causal Network Analysis for Quantitative Evaluation of Baseball Form by Video Analysis
Authors:

Takeshi Tanaka and Norio Gouda
Abstract: In professional sports, IT technology has been introduced to strengthen players, and in the baseball industry, where there is a large number of athletes, data analysis is expected to improve training efficiency, even for amateurs. However, the use of IT has not been widespread in the past due to the need for special equipment for measurement and the difficulty of interpreting data. In this study, we proposed a technique for quantifying the interlocking nature of players' forms using the transfer entropy from time-series data of players' skeletal coordinates obtained by image recognition to intuitively visualize the characteristics of players' forms using only video. As a result of evaluating players' hitting form using the proposed technique, we confirmed that the transfer entropy significantly changed in the target region when players were conscious of improvement, and we obtained a prospect for the practical application of form analysis using video.
Download
Paper Nr: 62
Title:

Impact Distance Detection in Tennis Forehand by an Inertial System
Authors:

Lucio Caprioli, Francesca Campoli, Saeid Edriss, Elvira Padua, Amani Najlaoui, Cristian Romagnoli, Giuseppe Annino and Vincenzo Bonaiuto
Abstract: Assessing the distance to the ball in the tennis forehand is fundamental. In this context, a non-invasive assessment system can help technicians even more in amateur tennis, where players who are still unaware of the act need continuous feedback. Three amateur tennis players with an average of 4 years of playing experience were recruited. The subjects wore a sensorized chest strap with an inertial unit and received two sets of 10 balls each. Two action cameras captured 20 forehands of each player from lateral and rear perspectives, aligned about 6m from the point of impact. Video analysis was conducted to identify the anteroposterior and lateral distance of the ball at the point of impact from the longitudinal axis coincident with the first toe of the nondominant foot. Pearson's correlation between distance and trunk inclination during the impact phase was investigated, and a strong correlation was found for all the subjects. This prompts us to consider the potential of a sensorized chest strap to assess the individual optimal distance from the ball in the forehand of tennis amateurs. Subsequent studies are needed to develop the system's full potential, expand the number of subjects, and examine all the fundamentals of the game.
Download