icSPORTS 2022 Abstracts

Area 1 - Computer Systems in Sports

Full Papers

Paper Nr:	4
Title:	Development of a Serious Game to Improve Decision-making Skills of Martial Arts Referees
Authors:	André Salmhofer, Lucian Gutica-Florescu, Dominik Hoelbling, Roland Breiteneder, Rene Baranyi and Thomas Grechenig
Abstract:	While sports referees need to cover a wide spectrum of demands depending on the characteristics of the judged sport, the outstanding responsibility they are associated with, is the task of decision-making. The focus of martial arts referees lies in perception and cognitive processing to detect, categorize and evaluate fast-moving techniques performed within a short period. To accumulate the training intensity required to reach expert level, recent research suggests complementing competitive experience with a video-based training approach. By combining the benefits of video-based training with motivational game elements, the study aimed to develop a video-based serious game to train intuitive decision-making processes of martial arts referees through immediate feedback. The training platform called JudgED comprises two modules: (a) a serious game to train decision-making processes and (b) a content and administration interface to manage, prepare, annotate and augment the video content used in the serious game. To evaluate the effectiveness of the serious game, a method to measure the players’ decision accuracy and reaction time is proposed.
Download

Paper Nr:	17
Title:	Using LSTM Networks and Future Gradient Values to Forecast Heart Rate in Biking
Authors:	Henry Gilbert, Jules White, Quchen Fu and Douglas C. Schmidt
Abstract:	Heart Rate prediction in cycling potentially allows for more effective and optimized training for a given individual. Utilizing a combination of feature engineering and hybrid Long Short-Term Memory (LSTM) models, this paper provides two research contributions. First, it provides an LSTM model architecture that accurately forecasts the heart rate of a bike rider up to 10 minutes into the future when given the future gradient values of the course. Second, it presents a novel model success metric optimized for deriving a model’s accuracy to predict heart rate while an athlete is zone training. These contributions provide the foundations for other applications, such as optimized zone training and offline reinforcement models to learn fatigue embeddings.
Download

Short Papers

Paper Nr:	6
Title:	Activity Trackers: Comparing Athlete Runners versus Health Runners through a Dedicated Technology Acceptance Model
Authors:	Ricardo Sol and Karolina Baras
Abstract:	The conducted study seeks to learn if, why and how two different groups of Activity Trackers users, Athletes and Health Runners, are utilizing these devices for their self-quantification. The study is based on the content analysis of 20 semi-structured interviews, 10 of which were with Athletes. To achieve its goals, the authors use a model based on the Technology Acceptance Model (TAM), a widely adopted technology acceptance theory. Amongst our findings, the construct Perceived Ease of Use showed that Athletes find it hard to program the settings for their training and Health Runners expressed that there is too much information involved. This paper contributes by showing that an all-purpose interface is not suitable and offers new knowledge for methodological discussions as it is, to the best of our knowledge, the first qualitative study to employ a TAM like model in order to qualitatively interpret the use of Activity Trackers.
Download

Paper Nr:	14
Title:	Body Movement Recognition System using Deep Learning: An Exploratory Study
Authors:	António Saleiro, Marcos Costa, Jorge Ribeiro and Bruno Silva
Abstract:	This paper presents an exploratory work using Deep Convolutional Neural Networks (CNN) in the context of detecting correct human movements during sports exercises, in particular specific physical exercises with the capability to perform the “5 times Sit-to-Stand Test” following the guideline of the Delaware university which is used in several settings to evaluate older adults’ functionality. It is used a supervised learning model implemented with an opensource approach for the detection of the human body position with the calculation aspect and angles of the limb movements (e.g. the angle of the knee) in order to recognize if the exercises were or not well executed. Results have shown that, under normal operating conditions, an accuracy in the exercises detection around 80% using the CNN Mobilenet architecture. We conclude that with the implementation of the Posenet algorithm using Tensorflow via ML5.js we achieved results of correct exercises accuracy between 53% and 79% for exercises involving the lower limbs and accuracy around 45% to the upper limbs. Due the promising results achieved, we intent in the future to extend the work with complementary approaches of human body pose estimation to improve the training strategy and different CNN network architectures to perform the recognition accuracy in the applicability context of correct execution exercises in elderly population physical activity scenarios.
Download

Paper Nr:	16
Title:	The Elusive Features of Success in Soccer Passes: A Machine Learning Perspective
Authors:	Hugo Muacho, Ricardo Ribeiro and Rui J. Lopes
Abstract:	Machine learning has in recent years been increasingly used in the soccer realm. This paper focuses on investigating the factors influencing pass success, a chief element in team performance. Decision tree techniques are used aiming to identify which features are the most important in pass success. This process is applied to a data set of 13 matches of the men’s French “Ligue 1”. Two experiments are conducted using different feature sets: one containing the positional data and Voronoi area off all players, the second considering only the ball carrier and closest teammates and opponents. The results obtained with the first feature set indicate that the relative importance of features is match dependent and somehow related to teams’ formation and players’ tactical mission. The second feature set, being more directly related to the passing process, provided a more consistent ranking of features. Features related to the interaction with the opponent standout. Low precision and recall values show that the features and factors leading to pass success are in fact elusive.
Download

Paper Nr:	18
Title:	Computational Fluid Dynamics to Reach a High-fidelity Simulator of Performance in Rowing
Authors:	Alban Leroyer and Sophie Barré
Abstract:	The massive growth of computational power and the advances of the numerical models make the use of numerical simulations to help analysing and improving sport performance achievable. However, it is still challenging because the physical configurations generally involved complex coupled problems and because human is part of the system. Furthermore, elite athletes already operate near an optimal point. As a consequence, the modelization of all the phenomena that come into play has to be accurate enough to be useful and relevant when the objective is to analyse interactions and to give reliable trends while varying some parameters. The case of rowing is presented here, through the development of SPRing (Simulator of Performance in Rowing), a high-fidelity simulator of the global system "boat-oars-rower(s)" coupled with the resolution of the Reynolds-Averaged Navier-Stokes equations to provide fluid forces acting on it.
Download

Paper Nr:	20
Title:	Novel Approach for a Hybrid Cushioning System in Running Shoes based on Halbach Arrays
Authors:	Philipp Kornfeind and Arnold Baca
Abstract:	A novel design approach making use of purely elastic magnet modules based on Halbach arrays for hybrid midsole constructions is proposed. Recovery of the surrounding midsole material after compression may thus be supported. In order to overcome challenges induced by the properties of ferromagnetic materials specific design principles had to be developed and implemented. In particular, mechanical guidance of the magnetic elements in vertical direction had to be ensured. FEM simulation has been applied in order to identify practicable geometrical arrangements of the magnets. Sufficient guiding accuracy was achieved by developing an effective structure utilizing dual Halbach arrays pivoted at a flat angle to one another using a common axis of rotation. Prototypes of midsoles and running shoes which utilize the novel technique have been manufactured. Validation tests revealed significantly higher stiffness of the modified midsoles and a reduced spring deflection. Biomechanical tests of the modified running shoes are in progress.
Download

Paper Nr:	13
Title:	A Preliminary Investigation into the Factors Impacting Tyre Failure in Wheelchair Basketball
Authors:	Rajtilak Kapoor, Hugo Espinosa and David Rowlands
Abstract:	Wheelchair sports are increasingly gaining popularity amongst athletes and spectators alike. This report outlines work carried out investigating the physical influences on the equipment used for Wheelchair Basketball. The preliminary experimental study explores the degradation of wheelchair tyres through the analysis of temperature, inflation pressure, acceleration, gameplay, individual playstyle, and strategy. With the aid of pressure monitoring and inertial measurement technology the study investigated the impact of player action on the wheelchairs. The experimental study was carried out in two phases employing the first phase of testing to provide the basis of the second phase. Through the first phase it was observed that the different temperature modalities did not have a significant impact on the degradation. The second phase focused on pressure observations and inertial sensor data paired with video. Through this process, the effect of player action was observed to have the greatest impact on the tyre degradation. Aggressive playstyles characterised by heavy impacts, and sharp decelerations were shown to play a significant role in reducing equipment reliability.
Download

Paper Nr:	19
Title:	Tracking Data Visual Representations for Sports Broadcasting Enrichment
Authors:	Murilo Couceiro, Inês Rito Lima, Alexandre Ulisses, Tiago Mendes-Neves and João Mendes-Moreira
Abstract:	The broadcast of audio-video sports content is a field with increasingly larger audiences demanding higher quality content and involvement. This growth creates the necessity to develop more content to engage the users and keep this trend. Otherwise, it may stall or even diminish. Therefore, enhancing the user experience, engagement, and involvement during live sports event broadcasts is of utmost importance. This paper proposes a solution to extract event’s information from video, resorting to Computer Vision techniques and Deep Learning algorithms. More specifically, the project encompassed the definition and implementation of field registration, object detection and tracking tasks. Focusing on football sports events, a novel dataset combining several video sources was created and used for analysis and metadata extraction. In particular, the proposed solution can detect and track players with acceptable precision using state-of-the-art methods, like YOLOv5 and DeepSORT. Furthermore, resorting to unsupervised learning techniques, the system provides team segmentation based on the colour of the players’ kits. A series of visual representations regarding the players’ movements on the field enables broadcast enrichment and increased user experience. The presented solution is framed in the H2020 DataCloud project and will be deployed in a cloud environment simplifying its access and utilisation.
Download

Area 2 - Health and Support Technology

Full Papers

Paper Nr:	2
Title:	On-demand Electric Thrust Assistance for Muscle-powered Watercraft: PADDELEC
Authors:	René Budich, Stephan Zipser, Jakob Doblaski and Jonas Seidel
Abstract:	In the project PADDELEC, the implementation of an intelligent paddle (iPaddle) for controlling an electric assisted canoe was realized. This paddle is able to record the paddle force FB of the blade with the help of measurement technology integrated in the paddle shaft. The recording of the paddle stroke is used as a control variable for operating an electric auxiliary drive in the boat. This is intended to provide on-demand thrust assistance to the athlete, analogous to an electrified bicycle ( pedelec). Existing drive solutions on motorized boats offer the athlete the option of firmly specifying the desired thrust via a thrust lever or tiller. The paddelec has the aim to provide intelligent and dynamic thrust assistance to the athlete as needed. This is intended to preserve a natural driving feel, despite the auxiliary drive. A simulation model of the longitudinal dynamics of canoes, was developed to investigate various assistance strategies. The models were validated by extensive real driving tests. For this purpose, special hardware and software tools had to be created, which will be further developed in the future. With the simulation and the evaluation of the practical testing, it could be shown that thrust support for canoes can be very useful. The correct support strategy and support performance can vary greatly depending on which water conditions the boat will use and the sporting activity, so there is still a need for more research in this area.
Download

Paper Nr:	9
Title:	Machine Learning for Fatigue Detection using Fitbit Fitness Trackers
Authors:	Erik J. Husom, Rustem Dautov, Adela-Aniela Nedisan, Fotis Gonidis, Spyridon Papatzelos and Nikolaos Malamas
Abstract:	Fatigue can be a pre-cursor to many illnesses and injuries, and cause fatal work-related incidents. Fatigue detection has been traditionally performed in lab conditions with stationary medical-grade diagnostics equipment for electroencephalography making it impractical for many in-field scenarios. More recently, the ubiquitous use of wearable sensor-enabled technologies in sports, everyday life or fieldwork has enabled collecting large amounts of physiological information. According to recent studies, the collected biomarkers related to sleep, physical activity or heart rate have proven to be in correlation with fatigue, making it a natural fit for applying automated data analysis using Machine Learning. Accordingly, this paper presents our novel Machine Learning-driven approach to fatigue detection using biomarkers collected by general-purpose wearable fitness trackers. The developed method can successfully predict fatigue symptoms among target users, and the overall methodology can be further extended to other diagnostics scenarios which rely on collected wearable data.
Download

Short Papers

Paper Nr:	1
Title:	Does Kinesio Taping Functional Correction Technique Affect Walking Plantar Pressures?
Authors:	Dalibor Kiseljak, Filip Bolčević and Vladimir Medved
Abstract:	The use of the Kinesio Taping (KT) method is widespread in sports, but current research on the ability of KT to alter foot biomechanics is limited. The aim of this study was to examine the acute effects of KT on walking plantar pressures. Kinesio Taping Functional Correction Technique was applied on the transverse arch of the foot. Twenty-two young healthy male adults, aged between 19 and 33 voluntarily participated in this study. Plantar pressures for three functional segments (forefoot, midfoot and hindfoot) were recorded during walking (with and without KT) by pedobarographic measurement using FDM 1.5 pressure measuring device. The results show that KT Functional Correction Technique has an effect in reducing walking plantar pressures. It was shown that maximal plantar pressures at the forefoot and hindfoot significantly (P < 0.05) decreased by 1.51 N/cm2 and 0.9 N/cm2 respectively, when KT was applied to the foot during walking. KT has the potential for primary and secondary prevention of pathological manifestations in the mentioned area, but also in the holistic postural context. Further research is needed to investigate clinical significance of KT.
Download

Paper Nr:	26
Title:	Pes Planus Plantaris (Flat Foot) Decreases Postural Stability of Basketball Student-athletes through Ground Reaction Force Vector (vGRF)
Authors:	Def Primal, Sasanty Kusumaningtyas and Ermita I. Ibrahim
Abstract:	Purpose: The main objective of this study is to determine the pes planus plantaris condition can contribute to the disturbance of postural stability on basketball athlete's in static and dynamics activities. Methods: This cross-sectional quantitative analytical retrospective study on 47 subjects of basketball student-athletes were identified the foot arch index by footprint extensive area, and AMTI Force flat-form (force plate) were determined their postural stability. Subjects were conducted in three activities (static, dynamic vertical jump, and dynamic loading response) for GRF resultant force vectors towards the vertical plane of body mass (W). Results: Analytical result obtained 80.9% of subjects had pes planus plantaris. It shows no significant differences in pes planus plantaris incidence in both sexes subject (p>0.005), however, there are differences in athlete’s exercise period aspect. Athlete students who have practiced strictly more than 4 years experienced over 50% for pes planus plantaris, furthermore, long period exercise were believed stimulating pes planus. The average value of GRF forces vectors of pes planus plantaris subjects on three different basketball movements shows a significant correlation to postural stability. Conclusions: Pes planus plantaris affected almost basketball athlete regarding the length and intensity of exercise performed. The condition significantly contribute to postural stability disturbance on a static condition, dynamic vertical jump, and dynamic vertical jump loading response.
Download

Area 3 - Sport Performance and Support Technology

Full Papers

Paper Nr:	22
Title:	Training with a Pneumatic Assist Suit to Generate Lower-Body Twisting during the Forehand Swing in Table Tennis
Authors:	Akihiko Kashiwagi, Hayato Nakamura, Satoshi Nishikawa and Kazuo Kiguchi
Abstract:	Movement form is important in sports. However, self-learning of the form has a possibility that people will acquire a form which puts a burden on one part of the body or a form which cannot exert sufficient power. Although there is also a way to have an instructor, people do not always receive the instruction of the form. Therefore, it is useful for sports training to develop a device which allows participants to acquire a form suited for each sport. Existing research about support in sports with a racket does not pay much attention to the lower body. In this research, we developed an assist suit that assists the lower body in executing a forehand swing in table tennis with lower-body twisting. Using this suit for beginners of table tennis, we conducted experiments under four conditions:(1) "No wear (before)", (2) "Without assist", (3) "With assist", and (4) "No wear (after)". As a result of Tukey analysis within each participant, the range of movement of the lower body is statistically increased by the assist suit and there are individual differences in whether to acquire a swing with twisting the lower body.
Download

Short Papers

Paper Nr:	7
Title:	Can Health Status and Lifestyle Indicators Predict Amateur Soccer Players Performance Level? A Preliminary Study
Authors:	Beatrice De Lazzari, Giuseppe Vannozzi, Federico Caramia, Filippo Lupi, Paolo Salvatore and Valentina Camomilla
Abstract:	Introduction: Soccer is one of the most popular sports in the world. To determine the performance potential of an athlete, various tests are typically performed in elite athletes, but not in amateur ones. Aim: To evaluate if and which health status and lifestyle indicators can be useful predictors of physical performance level in amateur soccer players. Methods: A group of 32 male subjects (age: 32±12 years, mass: 77±10 kg, stature: 1.78±0.06 m) voluntarily participated in the study. To assess their functional capacities, five in-field tests were conducted, while as an anamnesis sheet, a questionnaire was developed that investigated: body mass index (BMI), age, physical activity level, lifestyle, alcohol consumption and smoking habits, sports career, occurring injuries, and medical history. A stepwise backward regression was then conducted. Results: A significant R2=0.722 was found between the questionnaire outputs and the physical tests, using only six of the nine investigated indicators. Conclusions: With a simple questionnaire, an estimate of amateur athletes’ physical performance can be obtained. Prospectively, a wider dataset, including women, will allow for the definition of a synthetic biometric index.
Download

Paper Nr:	21
Title:	Assessing Psychomotor Abilities in Handcyclists using Computerized Tests: An Initial Study
Authors:	Wojciech Pasko, Justyna Dydek, Janusz Zielinski, Tomasz Hulewicz, Maciej Śliż, Bartosz Dziadek and Krzysztof Przednowek
Abstract:	Psychomotor abilities play an important role in any sport, they affect the speed of decisions made in a given situation, and thus can have a direct impact on the final outcome of sports competition. The study included ten disabled handcyclists who represented Poland internationally in the H2, H3, H4 and H5 categories. The control group consisted of non-disabled 5th year physical education students at the University of Rzeszow. The research methods were psychomotor tests using computer techniques in the Test2Drive system. For the evaluation, the tests performed were: simple reaction time (SIRT), complex reaction time (CHORT), hand-eye coordination (HECOR) and spatial orientation (SPANT). Reaction time (RT), motor time (MT) and correct responses (c.r.) were analyzed. The group of disabled cyclists is characterized by better motor time in each test while no statistical significance was shown. A better reaction time in each psychomotor test was achieved by the control group and the differences are statistically significant. Better motor time may suggest that cycling training has a positive effect on psychomotor abilities.
Download