No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Machine and deep learning for sport-specific movement recognition: a systematic review of model development and performance
Abstract
Objective assessment of an athlete's performance is of importance in elite sports to facilitate detailed analysis. The implementation of automated detection and recognition of sport-specific movements overcomes the limitations associated with manual performance analysis methods. The object of this study was to systematically review the literature on machine and deep learning for sport-specific movement recognition using inertial measurement unit (IMU) and, or computer vision data inputs. A search of multiple databases was undertaken. Included studies must have investigated a sport-specific movement and analysed via machine or deep learning methods for model development. A total of 52 studies met the inclusion and exclusion criteria. Data pre-processing, processing, model development and evaluation methods varied across the studies. Model development for movement recognition were predominantly undertaken using supervised classification approaches. A kernel form of the Support Vector Machine algorithm was used in 53% of IMU and 50% of vision-based studies. Twelve studies used a deep learning method as a form of Convolutional Neural Network algorithm and one study also adopted a Long Short Term Memory architecture in their model. The adaptation of experimental set-up, data pre-processing, and model development methods are best considered in relation to the characteristics of the targeted sports movement(s).
... Vision is much more than seeing clearly, it is interconnected visual skills which affects performance. Just as exercises and drills can improve speed and strength, it can also improve the athlete's visual fitness and visual accuracy (3). In sport, vision may affect athlete's performance, including visual clarity, athletic performance (the ability to perform specific tasks) and information processing. ...
... The science of improving visual skills to assist athletes in achieving the highest performance levels are becoming increasingly important in training many sports. Sport science performance analysis has undergone considerable changes recently, mainly due to the increased access and application of improved technology in computer science (3). Vision tests and sports training can help athletes determine the function of the eyes, beyond a basic ability to clearly see the letters and objects on standard eye charts (4). ...
... Athletes can devote thousands of hours of physical training to improve their physical fitness; however, if vision or visual processing ability is insufficient, physical training may not be optimized and athletic performance may be affected (6). Athletes who make full use of the potential of the visual system will gain optimal levels of performance (3). ...
In ball sports such as soccer, the visual system is critical in guiding a player’s search for crucial information that underpins skillful behavior, which requires the incorporation of all of the relevant information in the environment in order to make successful decisions under pressure. However, vision in sport, and focusing on the specific visual skills required to be successful in a particular sport has largely been a practice ignored by experts and coaches as being an essential component of athletic performance. This is the first attempt to summarize and compile the necessary visual skills for soccer. This review’s evidence suggests that, while current research still tends to focus on visual
skills as a whole, there is a need to streamline this focus to the necessities of a particular sport. Furthermore, in identifying the visual skills essential for soccer, it allows for the effective training and testing of these skills, as well as for talent identification.
... Over the past decade, machine learning modeling techniques have been developed and used to determine PA type from PA and sport-based monitoring devices (Atallah et al., 2011;Bonomi et al., 2009;Cust et al., 2019). While machine learning offers a unique opportunity to examine PA type, current machine learning models mainly focus on determining energy expenditure or PA types related to general health (e.g., sitting and walking) or specialized sport activities (e.g., quality of a soccer pass; Cust et al., 2019). ...
... Over the past decade, machine learning modeling techniques have been developed and used to determine PA type from PA and sport-based monitoring devices (Atallah et al., 2011;Bonomi et al., 2009;Cust et al., 2019). While machine learning offers a unique opportunity to examine PA type, current machine learning models mainly focus on determining energy expenditure or PA types related to general health (e.g., sitting and walking) or specialized sport activities (e.g., quality of a soccer pass; Cust et al., 2019). Many models also combine information from monitors worn in multiple locations and/or from videos captured during the session, both of which require significant burden to the athlete and/or coaching staff (Ahmadi, 2015;Cust et al., 2019). ...
... While machine learning offers a unique opportunity to examine PA type, current machine learning models mainly focus on determining energy expenditure or PA types related to general health (e.g., sitting and walking) or specialized sport activities (e.g., quality of a soccer pass; Cust et al., 2019). Many models also combine information from monitors worn in multiple locations and/or from videos captured during the session, both of which require significant burden to the athlete and/or coaching staff (Ahmadi, 2015;Cust et al., 2019). By contrast, recognition of PA types related to functional status or activities common across many sports and with a single body-worn device have received less attention. ...
A universal approach to characterizing sport-related physical activity (PA) types in sport settings does not yet exist. Young adults (n = 30), 19–33 years, engaged in a 15-min activity session, performing warm-ups, 3-on-3 soccer, and 3-on-3 basketball. Videos were recorded and manually coded as criterion PA types (walking, running, jumping, rapid lateral movements). Participants wore an accelerometer on their right hip. Multiple machine learning models were developed and compared for predicting PA type. Most models underestimated time spent completing the activities performed least commonly. Point estimates for percent agreement, sensitivity, specificity, F-scores, and kappa were similar across models, with Hidden Markov Models (HMMs) being best at classifying rare events. Models detected activity type during sport-related movements with modest accuracy (kappas ≤ .40). Given the better performance of HMMs, incorporating the temporal nature of sport-related activities is important for improving sport-related PA classification.
... In the literature, a great variety of sport-specific activity recognition systems exists, which can be used for fine-grained, athlete-specific monitoring [6]. To detect actions reliably, a sufficient amount of training data needs to be recorded, which is not only cumbersome but sometimes not possible at all, especially in marginal sports. ...
... However, those come with the disadvantages of being expensive, causing privacy issues, and having to be mounted at specific positions, which may not be feasible for minor sports outside of huge stadiums [10]. On the other hand, sensorbased systems using IMUs like accelerometers, gyroscopes, and magnetometers have been adopted in the research community [6]. Historically, accurate sensors had to be custom-built for each application. ...
... Smaller, cheaper, and better availability of sensors have made it possible to analyze numerous different sports and movements in recent years [9]. Cust et al. [6] provide a broad overview of machine and deep learning for sportspecific movement recognition. Their review includes a comparison of several sensor-and vision-based solutions for HAR in sports. ...
In human activity recognition har(human activity recognition (HAR)), activities are automatically recognized and classified from a continuous stream of input sensor data. Although the scientific community has developed multiple approaches for various sports in recent years, marginal sports are rarely considered. These approaches cannot directly be applied to marginal sports, where available data are sparse and costly to acquire. Thus, we recorded and annotated inertial measurement unit (IMU) data containing different types of Ultimate Frisbee throws to investigate whether Convolutional Neural Network (CNNs) and transfer learning can solve this. The relevant actions were automatically detected and were classified using a CNN. The proposed pipeline reaches an accuracy of 66.6%, distinguishing between nine different fine-grained classes. For the classification of the three basic throwing techniques, we achieve an accuracy of 89.9%. Furthermore, the results were compared to a transfer learning-based approach using a beach volleyball dataset as the source. Even if transfer learning could not improve the classification accuracy, the training time was significantly reduced. Finally, the effect of transfer learning on a reduced dataset, i.e., without data augmentations, is analyzed. While having the same number of training subjects, using the pre-trained weights improves the generalization capabilities of the network, i.e., increasing the accuracy and F1 score. This shows that transfer learning can be beneficial, especially when dealing with small datasets, as in marginal sports, and therefore, can improve the tracking of marginal sports.
... However, seeing the player's position and velocity in relation to all other players in the field could lead to interesting insight in the player's intention, fitness or overall performance. Moreover, kinematic features can be processed by machine learning algorithms in order to automatically interpret the current performance of an athlete to a greater extent [43]. Sports-or action-specific models are trained with previously obtained, known data (so-called 'training data'). ...
... In the field of machine learning, Cust et al. [43] published a systematic review on the automated recognition of sports-specific movements. The review provides an overview of the field of supervised machine learning with additional consideration of the comparison between IMMU and visual data, data fusion approaches and the perspective of deep learning and unsupervised learning. ...
... In the field of machine learning in sports, mainly supervised and unsupervised learning are of interest [43]. Supervised learning is applied to known scenarios that allow for (manual) data annotation, whereas unsupervised learning can be applied to scenarios of data with (partially) unknown content. ...
Sports analytics research has major impact on the development of innovative training methods and the broadcast of sports events. This dissertation provides algorithms for both kinematic analysis and performance interpretation based on unobtrusively obtained measurements from wearable sensors. Its main focus is set on the processing of 3D-orientation features and the exploration of their potential for sports analytics. The proposed algorithms are described and evaluated in five exemplary sports. In scuba diving, rowing and ski jumping, the 3D-orientation of the body/boat/skis is determined and further processed to analyze and visualize the motion behavior. In snowboarding and skateboarding, the board orientation is calculated and processed for motion visualization and machine learning. Board sport tricks are automatically detected and subsequently classified for trick category and type. The methods of this work were already partially applied for TV broadcast of international competitions (e.g., Olympics 2018). Additionally, they can support sports science research for establishing thorough investigations and innovative training methods.
... In competitive and professional sports, all athletes are monitored in nearly every game and, if possible, also during training. The monitoring provides data that can be analysed to further improve the performance of individual athletes or the team, but it can also deliver information about opposition teams, their tactics and strategy, strength and weaknesses, etc. [1][2][3][4][5]. As the amount of available data is too large to be processed efficiently by coaches and analysts, the state of the art in the analysis of such data comprises a mixture of computer-aided and human analysis and evaluation [2,[5][6][7][8]. ...
... The monitoring provides data that can be analysed to further improve the performance of individual athletes or the team, but it can also deliver information about opposition teams, their tactics and strategy, strength and weaknesses, etc. [1][2][3][4][5]. As the amount of available data is too large to be processed efficiently by coaches and analysts, the state of the art in the analysis of such data comprises a mixture of computer-aided and human analysis and evaluation [2,[5][6][7][8]. The computer-aided part of the analysis is mostly based on modern algorithms, e.g., methods of machine learning [9][10][11][12][13], though before any analysis can be carried out, the data has to be gathered. ...
In order to train receivers in American football in a targeted and individual manner, the strengths and weaknesses of the athletes must be evaluated precisely. As human resources are limited, it is beneficial to do it in an automated way. Automated passing machines are already given, therefore the motivation is to design a computer-based system that records and automatically evaluates the athlete’s catch attempts. The most fundamental evaluation would be whether the athlete has caught the pass successfully or not. An experiment was carried out to gain data about catch attempts that potentially contain information about the outcome of such. The experiment used a fully automated passing machine which can release passes on command. After a pass was released, an audio and a video sequence of the specific catch attempt was recorded. For this purpose, an audio-visual recording system was developed which was integrated into the passing machine. This system is used to create an audio and video dataset in the amount of 2276 recorded catch attempts. A Convolutional Neural Network (CNN) is used for feature extraction with downstream Long Short-Term Memory (LSTM) to classify the video data. Classification of the audio data is performed using a one-dimensional CNN. With the chosen neural network architecture, an accuracy of 92.19% was achieved in detecting whether a pass had been caught or not. The feasibility for automatic classification of catch attempts during automated catch training is confirmed with this result.
... Measuring sports movement during training and competition allows monitoring athletes' performance and their risk of injury (Camomilla, Bergamini, Fantozzi, & Vannozzi, 2018;Cust, Sweeting, Ball, & Robertson, 2019). Performance monitoring is relevant to assess motor capacity and physical demand, as well as to analyze technique and how technique impacts performance (Camomilla et al., 2018). ...
... Machine learning is a powerful tool where the recorded data are used to make inferences about the data themselves (Bishop, 2006). Supervised learning, unsupervised learning, and reinforcement learning are increasingly investigated in human movement analysis, for example, to classify normal and pathological gait, to map IMU data to biomechanical variables, to discover or cluster movement patterns, or to learn controllers that drive biomechanical models (Cust et al., 2019;Ferber, Osis, Hicks, & Fig. 1. Overview of "in the wild" movement recording and analysis using machine learning. ...
Recent advances in wearable sensing and machine learning have created ample opportunities for “in the wild” movement analysis in sports, since the combination of both enables real-time feedback to be provided to athletes and coaches, as well as long-term monitoring of movements. The potential for real-time feedback is useful for performance enhancement or technique analysis, and can be achieved by training efficient models and implementing them on dedicated hardware. Long-term monitoring of movement can be used for injury prevention, among others. Such applications are often enabled by training a machine learned model from large datasets that have been collected using wearable sensors. Therefore, in this perspective paper, we provide an overview of approaches for studies that aim to analyze sports movement “in the wild” using wearable sensors and machine learning. First, we discuss how a measurement protocol can be set up by answering six questions. Then, we discuss the benefits and pitfalls and provide recommendations for effective training of machine learning models from movement data, focusing on data pre-processing, feature calculation, and model selection and tuning. Finally, we highlight two application domains where “in the wild” data recording was combined with machine learning for injury prevention and technique analysis, respectively.
... Monitoring physical exercises is a well studied topic in mobile, wearable and ubiquitous human activity recognition [21,59,70]. It can help athletes to track their performances, and the broader consumers to document their activity levels and achieve their fitness goals. ...
... Wearable motion sensors such as IMUs have been extensively studied in the sports activity recognition field especially in categorizing exercises [6,21,37,40,52]. However, studies with IMU-based systems have not yet provided such performance on body weight exercises. ...
While sports activity recognition is a well studied subject in mobile, wearable and ubiquitous computing, work to date mostly focuses on recognition and counting of specific exercise types. Quality assessment is a much more difficult problem with significantly less published results. In this work, we present Quali-Mat: a method for evaluating the quality of execution (QoE) in exercises using a smart sports mat that can measure the dynamic pressure profiles during full-body, body-weight exercises. As an example, our system not only recognizes that the user is doing push-ups, but also distinguishes 5 subtly different types of push-ups, each of which (according to sports science literature and professional trainers) has a different effect on different muscle groups. We have investigated various machine learning algorithms targeting the specific type of spatio-temporal data produced by the pressure mat system. We demonstrate that computationally efficient, yet effective Conv3D model outperforms more complex state-of-the-art options such as transfer learning from the image domain. The approach is validated through an experiment designed to cover 47 quantifiable variants of 9 basic exercises with 12 participants. Overall, the model can categorize 9 exercises with 98.6% accuracy / 98.6% F1 score, and 47 QoE variants with 67.3% accuracy / 68.1% F1 score. Through extensive discussions with both the experiment results and practical sports considerations, our approach can be used for not only precisely recognizing the type of exercises, but also quantifying the workout quality of execution on a fine time granularity. We also make the Quali-Mat data set available to the community to encourage further research in the area.
... To the best of our knowledge, there is no such literature review for sports technology overall available yet. However, there are several literature reviews that cover sub-domains of sports technology in different areas such as training (Rajšp and Fister, 2020;Simim et al., 2017), performance (van den Berg et al., 2020), rehabilitation (Cooper and Cooper, 2019), injury prevention (Tjønndal et al., 2022), health (Alqahtani et al., 2021;Duvall et al., 2021), motion capture (Jenny et al., 2017;van der Kruk and Reijne, 2018), biomechanics (Taborri et al., 2020), wearables (Aroganam et al., 2019;Mencarini et al., 2019), heart rate (Jiménez Morgan and Molina Mora, 2017), artificial intelligence (Bonidia et al., 2018;Ramkumar et al., 2021), machine learning (Cust et al., 2019), data collection (Owen et al., 2015), big data (Mataruna-Dos-Santos et al., 2020), virtual reality (Nor et al., 2020), additive manufacturing (Manoharan et al., 2013;Novak and Novak, 2020), innovation (Tjønndal, 2017), sport management (Tjønndal, 2016), sports fields (Straw et al., 2020), esports (Flegr and Schmidt, 2022;Reitman et al., 2019), sports betting (Winters and Derevensky, 2020), or ethics (Dyer, 2015). ...
... A threshold of 30 occurrences was selected, resulting in 100 keywords. (Andrienko et al., 2021;Anzer and Bauer, 2021;Bauer and Anzer, 2021), practice-enhancing and human enhancement technologies (Farrow, 2013;Faure et al., 2020;Miah, 2006;Miles et al., 2012), sportspecific movement recognition (Cust et al., 2019;Minh Dang et al., 2020;Wu et al., 2021), biomechanics (de Magalhaes et al., 2015, human motion capture and tracking systems for sport applications (Adesida et al., 2019;Barris and Button, 2008;Taborri et al., 2020;van der Kruk and Reijne, 2018), integrated technology and microtechnology sensors (Cummins et al., 2013;Dellaserra et al., 2014;Rago et al., 2020;Whitehead et al., 2018), or ubiquitous computing in sports (Baca et al., 2009;Chi, 2008). ...
Rapid technological progress and digitalization have considerably changed the role of technology in sports in the past two decades. As the human limits of performance have been reached in many disciplines, reaching future limits will increasingly depend on technology. While this represents progress in how athletes train and compete, similar developments await sports managers in the way they lead sports organizations and sports consumers in the way they consume and engage with sports. Using the SportsTech Matrix (i.e., a framework to capture how different types of technologies provide solutions to different user groups in sports), we examine how technology will impact sports in the future. We present a Delphi-based prospective study with quantitative and qualitative assessments from 92 subject matter experts for six future projections and 35 non-Delphi prospective survey items. We find that, by 2030, technology will significantly impact all three user groups in sports: athletes, consumers, and managers. For athletes, experts anticipate technology to play a major role for sporting performance improvements. For consumers, the consumption of sports content will continue to change significantly. For management, new types of manager profiles in terms of backgrounds and skill sets would be desirable. We discuss two possible future scenarios: (1) a probable future and (2) a game changer. Our findings should provide relevant insights for decision-makers and other stakeholders in sports and raise promising directions for future research.
... They also incorporated some of the available datasets of different sports. • Cust et al. [7] presented a systematic review on machine learning and deep learning for sports-specific movement recognition using inertial measurement units and computer vision data. ...
... • Rana et al. [13] offered a thorough overview of the literature on the use of wearable inertial sensors for performance measurement in various sports. -Sport data mining [3] -- [7] -- [4] -Motion Capture [8] Soccer Ball Tracking [5] Soccer Player detection/tracking [6] -Availability of datasets for sports [9] -Content-Aware Analysis [10] -- The proposed survey mainly focuses on providing a proper and comprehensive survey of research carried out in computer vision-based sports video analysis for various applications such as detection and classification of players, tracking players or balls and predicting the trajectories of players or balls, recognizing the team's strategies, classifying various events on the sports field, etc. and in particular, establishing a pathway for nextgeneration research in the sports domain. The features of this review are: ...
Recent developments in video analysis of sports and computer vision techniques have achieved significant improvements to enable a variety of critical operations. To provide enhanced information, such as detailed complex analysis in sports such as soccer, basketball, cricket, and badminton, studies have focused mainly on computer vision techniques employed to carry out different tasks. This paper presents a comprehensive review of sports video analysis for various applications: high-level analysis such as detection and classification of players, tracking players or balls in sports and predicting the trajectories of players or balls, recognizing the team’s strategies, and classifying various events in sports. The paper further discusses published works in a variety of application-specific tasks related to sports and the present researcher’s views regarding them. Since there is a wide research scope in sports for deploying computer vision techniques in various sports, some of the publicly available datasets related to a particular sport have been discussed. This paper reviews detailed discussion on some of the artificial intelligence (AI) applications, GPU-based work-stations and embedded platforms in sports vision. Finally, this review identifies the research directions, probable challenges, and future trends in the area of visual recognition in sports.
... Interested readers may refer to any of the numerous surveys that are available. An example of such surveys is [15,16]. Furthermore, there are numerous excellent works in the parent field of activity recognition. ...
... Personal fitness monitoring can make use of exercise recognition, as presented in [5,6,11,16,43]. After developing and testing a system for gym exercise recognition and obtaining good results, the next logical question is, what can be its practical applications? ...
Automatic tracking and quantification of exercises not only helps in motivating people but also contributes towards improving health conditions. Weight training, in addition to aerobic exercises, is an important component of a balanced exercise program. Excellent trackers are available for aerobic exercises but, in contrast, tracking free weight exercises is still performed manually. This study presents the details of our data acquisition effort using a single chest-mounted tri-axial accelerometer, followed by a novel method for the recognition of a wide range of gym-based free weight exercises. Exercises are recognized using LSTM neural networks and the reported results confirm the feasibility of the proposed approach. We train and test several LSTM-based gym exercise recognition models. More specifically, in one set of experiments, we experiment with separate models, one for each muscle group. In another experiment, we develop a universal model for all exercises. We believe that the promising results will potentially contribute to the vision of an automated system for comprehensive monitoring and analysis of gym-based exercises and create a new experience for exercising by freeing the exerciser from manual record-keeping.
... Regarding the positional data that can be obtained from drone footage, a review of the literature shows that several different methods based on image processing and computer vision have been used to automatically track players in a variety of sports (Cai and Aggarwal, 1996;Araki et al., 2000;Pers and Kovačič, 2000;Needham and Boyle, 2001;Iwase and Saito, 2004;Di Salvo et al., 2006;Figueroa et al., 2006;Barris and Button, 2008;Barros et al., 2011). More recently, new methods based on deep learning approaches, like convolutional neural networks, have improved the recognition and tracking of players in field sports, reducing the need of an operator to correct the tracking of players (Stein et al., 2017;Thomas et al., 2017;Cust et al., 2018;Renò et al., 2018). However, these methods all rely on multiple fixed cameras, and none have yet made use of a single drone camera. ...
... These studies also failed to validate the accuracy and reliability of the positional data obtained from the drone footage. Consequently, we believe that the current state of the art in computer vision and deep learning allows for tracking players automatically and provide positional data to derive performance indicators based on drone-based video technology (Thomas et al., 2017;Cust et al., 2018;Liang et al., 2019;Lee et al., 2020). ...
Radio and video-based electronic performance and tracking systems (EPTS) for position detection are widely used in a variety of sports. In this paper, the authors introduce an innovative approach to video-based tracking that uses a single camera attached to a drone to capture an area of interest from a bird's eye view. This pilot validation study showcases several applications of this novel approach for the analysis of game and racket sports. To this end, the authors compared positional data retrieved from video footage recorded using a drone with positional data obtained from established radio-based systems in three different setups: a tennis match during training with the drone hovering at a height of 27 m, a small-sided soccer game with the drone at a height of 50 m, and an Ultimate Frisbee match with the drone at a height of 85 m. For each type of playing surface, clay (tennis) and grass (soccer and Ultimate), the drone-based system demonstrated acceptable static accuracy with root mean square errors of 0.02 m (clay) and 0.15 m (grass). The total distance measured using the drone-based system showed an absolute difference of 2.78% in Ultimate and 2.36% in soccer, when compared to an established GPS system and an absolute difference of 2.68% in tennis, when compared to a state-of-the-art LPS. The overall ICC value for consistency was 0.998. Further applications of a drone-based EPTS and the collected positional data in the context of performance analysis are discussed. Based on the findings of this pilot validation study, we conclude that drone-based position detection could serve as a promising alternative to existing EPTS but would benefit from further comparisons in dynamic settings and across different sports.
... The model can give coaching and medical staff tackle-specific measurements, in real-time, which can be used in injury prevention and rehabilitation strategies. Following on from this, (Cust et al. 2018) reviewed the ways that machine learning and AI can be used to classify certain movements in sport. ...
The Sports Analytics Market is growing rapidly, in 2020 it was valued at over $1 billion and is expected to reach over $5 billion by 2026. However, even with this level of growth the use of Artificial Intelligence (AI) techniques have yet to fully be explored. The sports analytics domain presents a number of significant computational challenges for AI and Machine Learning. In this thesis, we propose a number novel methods for analysing team sports data to help sports teams utilise AI to improve their strategic and tactical decision making. By doing so, we present a number of contributions to the AI and sports analytics communities. In particular, we present a model for the tactical decisions that are made in football and show how game theoretic techniques can be used to optimise these. We focus on both the short-term decisions made for individual games, as well as longer term decisions to maximise performance over a season. We show that we can increase a teams chances of winning individual games by 16.1% and can increase a teams mean expected finishing position by up to 35.6%. We also, introduce a new model for valuing the teamwork between players in sports teams by assessing the outcomes of chains of interactions between the players in a team. We then present a novel model for forming teams based on this value and maximise teamwork by assessing the overlapping pairs in a team. Our model is shown to better predict the real-world performance of teams by up to 46% compared to models that ignore inter-agent interactions. Finally, we show how we can use natural language processing techniques to improve the traditional statistical methods for prediction sports match outcomes. We use domain expert written articles from the media to train our models and we show that by incorporating the features learned from the text, we can boost the accuracy of the traditional statistical methods by 6.9%.
... One option for developing the evidence-based training is to apply innovative technologies, such as wearable systems and machine learning in coaching practice. Such an approach would help the elite athletes to activate their potential capabilities and stretch human biological boundaries [11][12][13]. ...
Developing real-time biomechanical feedback systems for in-field applications will transfer human motor skills’ learning/training from subjective (experience-based) to objective (science-based). The translation will greatly improve the efficiency of human motor skills’ learning and training. Such a translation is especially indispensable for the hammer-throw training which still relies on coaches’ experience/observation and has not seen a new world record since 1986. Therefore, we developed a wearable wireless sensor system combining with artificial intelligence for real-time biomechanical feedback training in hammer throw. A framework was devised for developing such practical wearable systems. A printed circuit board was designed to miniaturize the size of the wearable device, where an Arduino microcontroller, an XBee wireless communication module, an embedded load cell and two micro inertial measurement units (IMUs) could be inserted/connected onto the board. The load cell was for measuring the wire tension, while the two IMUs were for determining the vertical displacements of the wrists and the hip. After calibration, the device returned a mean relative error of 0.87% for the load cell and the accuracy of 6% for the IMUs. Further, two deep neural network models were built to estimate selected joint angles of upper and lower limbs related to limb coordination based on the IMUs’ measurements. The estimation errors for both models were within an acceptable range, i.e., approximately ±12° and ±4°, respectively, demonstrating strong correlation existed between the limb coordination and the IMUs’ measurements. The results of the current study suggest a remarkable novelty: the difficulty-to-measure human motor skills, especially in those sports involving high speed and complex motor skills, can be tracked by wearable sensors with neglect movement constraints to the athletes. Therefore, the application of artificial intelligence in a wearable system has shown great potential of establishing real-time biomechanical feedback training in various sports. To our best knowledge, this is the first practical research of combing wearables and machine learning to provide biomechanical feedback in hammer throw. Hopefully, more wearable biomechanical feedback systems integrating artificial intelligence would be developed in the future.
... Workers in the construction industry are often exposed to physically demanding manual tasks with a high degree of ergonomic risk [71,72]. The rapid development of motion sensors in the construction industry enables proactive accident prevention by reducing the number of dangerous actions that commonly occur [73]. The authors of the articles Experience, Productivity, and Musculoskeletal Injury among Masonry Workers [74], Data Fusion of [71] want to achieve consistent results. ...
Due to the increasingly high proportion of manual activities in production processes, there is a constant risk of musculoskeletal disorders or work-related injuries. The risk of these problems is exacerbated by the trend towards an ageing working population. European legislation is pressing for improved working conditions to eliminate the risks associated with health problems for workers. For this reason, the application of ergonomics in this field is growing. Musculoskeletal disorders, which are most often caused by inappropriate working postures, are a major problem. There are many methods for evaluating working postures. However, there is a high degree of subjectivity in the risk assessment. Motion capture kinematic suits can ensure the objectivity of the assessment. This article discusses research on ergonomics assessment using motion capture technology. A systematic literature search method was used for the research, beginning with the determination of the research procedure, through the definition of the research queries, to the formulation of the research itself to identify relevant sources. The study presents the most widely used methods for assessing the ergonomics of work positions using motion capture technology, their advantages, and disadvantages. It also follows the trend in the number of publications between 2010 and 2022 in countries where the topic is most frequently addressed and in the industries where motion capture technology is used for ergonomics assessment in general. The research showed that this approach is most often used in industry and logistics, and less frequently in healthcare and sport. The authors agree that the most frequently used ergonomics assessment methods are not complex enough to be used in combination with motion capture and that a combination of the two is needed. At the same time, this technology has become very important in the field of ergonomic evaluation of work positions, offering a higher degree of objectivity, or can be combined with the use of virtual reality, but the evaluation systems are still not error-free and there is a need for continuous improvement.
... Machine learning and deep learning are the new technology in the computer world. According to (Cust et al., 2019) indicate the positive affect of using machine learning and deep learning to enhance the sport activity. It is mean sport activity can enhance and practices with using the technology. ...
Objectives:Sports participation has an important role to play. Sports activities are change environment to another. Also, using information and communication technology such as distance learning has become a rapidly growing trend. Online delivery platform becomes very important methods for the graduate and undergraduate students to achieves the advanced degrees and knowledge.The aim of this study to investigate the impact of online learning on Sports participation during pandemic (COVID-19). Methods: This study usedthe previous studiesto investigate theuse of online learning on Sports participation during pandemic (COVID-19). This research uses different databased such as Science direct, Google Scholar, Scopus and other databases. Results: This study comeswith some obstacles that face student in Saudi Arabia to practise the sport activities during the pandemic (COVID-19). Those obstacles should consider by Saudi ministry of education.
... As can be seen in Table 1, the overall performance of the various approaches used in this paper is summarized. Compared with existing methods [20][21][22], our proposed model performs better with a smaller error for different poses and parameters. According to Table 1, all existing models provide the human body joint points and the lines between the joint points as the body posture. ...
The EAF (Ethiopian Athletics Federation) is the main governing body of Ethiopian athletics. All athletes must undergo extensive training in order to compete in the Olympic Games and international competitions. Coaches acknowledge the training activities that their athletes have completed during the training process. The training processes are subjected to scientific interventions, and the training actions are contrasted with standard actions. For better training session results, this intervention is crucial. Every action that the athletes take must be evaluated by the coaches to ensure that it satisfies the minimum requirements. The reason for this is that because individual athletes perform their actions quickly, the coach's eye may not produce accurate results because human activities are prone to mistakes. So, using a convolutional neural network, this paper designs and develops an algorithm for recognizing and intervening in the motion and gestures of long-distance runners. In the suggested algorithm, the athlete posture image's texture features and HSV features are first extracted, and then a dual-channel CNN is built. Separately extracted values for each characteristic are combined from the dual-channel network's output. Finally, the constructed image of the athlete's posture is estimated and built using the results passed from a fully connected CNN. This article conducts ablation studies, experimental testing, and comparative analysis on a variety of data. This experimental work thus demonstrates that the suggested algorithm produces better performance results.
... Stöve et al. [17] used IMUs to detect individual shots and passes of soccer players with machine learning. Cust et al. [18] provide an overview of model development and performance for machine and deep learning for movement recognition in sports. Best practices have been established in the literature regarding data aggregation, data cleaning, model selection, and other components [16]. ...
With tracking systems becoming more widespread in sports research and regular training and competitions, more data are available for sports analytics and performance prediction. We analyzed 2523 ski jumps from 205 athletes on five venues. For every jump, the dataset includes the 3D trajectory, 3D velocity, skis’ orientation, and metadata such as wind, starting gate, and ski jumping hill data. Using this dataset, we aimed to predict the expected jump length (xLength) inspired by the expected goals metric in soccer (xG). We evaluate the performance of a fully connected neural network, a convolutional neural network (CNN), a long short-term memory (LSTM), and a ResNet architecture to estimate the xLength. For the prediction of the jump length one second after take-off, we achieve a mean absolute error (MAE) of 5.3 m for the generalization to new athletes and an MAE of 5.9 m for the generalization to new ski jumping hills using ResNet architectures. Additionally, we investigated the influence of the input time after the take-off on the predictions’ accuracy. As expected, the MAE becomes smaller with longer inputs. Due to the real-time transmission of the sensor’s data, xLength can be updated during the flight phase and used in live TV broadcasting. xLength could also be used as an analysis tool for experts to quantify the quality of the take-off and flight phases.
... The measurement and analysis of data and the corresponding dashboard technology can be said to be a whole field on its own. This is discussed at more length elsewhere (Kokaram et al., 2006;Perin et al., 2018); existing literature covers architectures for such systems (Matejka et al., 2014;Brunauer et al., 2020;Renò et al., 2017); hardware for measurement and tracking (e.g., Van der Kruk and Reijne, 2018;Fuss, 2008;Cust et al., 2019;Düking et al., 2018); tools for analysing and modelling sports data (e.g., Brefeld et al., 2019); novel forms of visualising (Perin et al., 2018;Polk et al., 2014) and querying (Shao et al., 2016) the data, or the decision making that builds upon this data (e.g. Vales-Alonso et al., 2015López-Matencio et al., 2010) -although the latter is also often left to the coach or athlete. ...
... 13,22,23 Other studies 9,24 evaluated the convergent validity and test-retest reliability between GPS devices of the same model in which it was observed that there was a difference in the results for the same variable and for the same test, concluding that some GPS models could vary in results between themselves, thus making it essential to carry out reliability studies to guarantee the reproducibility of the results from the equipment. On the other hand, most studies were carried out in non-ecological environments, creating circuits which simulate real conditions of competition 13,20,25 and therefore it is interesting to note that our study compared Instat with real tracking data during official matches using a GPS device. This is due to the recurring signal loss from the engineering of certain soccer stadiums, especially those with metallic covers, as described in the literature, 10,18 making it impossible to use these data from some matches in our study. ...
Objective
To test the reliability between two instruments with different analysis mechanisms, either by GPS (model GPSPORTS®) or by video analysis (InStat For Players®), relating the results of total distance covered and distance at high speed ≥ 20km/h (Very High-Intensity Running Distance, VHIR) during official soccer matches.
Study Design
This is a methodological study. Data from 35 male professional soccer athletes from all tactical positions were included. Age 29.2 (± 4.8 years) and body fat 9.9 (± 1.7%), excluding goalkeepers (102 individual analyzes) were collected in official matches. In the data analysis, descriptive statistics procedures were used to characterize the sample and the intraclass correlation coefficient (ICC) was used to verify the agreement on the stability and internal consistency of the tests with 95% confidence intervals (CI).
Results
The ICC in the case of the total distance traveled variable was significant 0,914 (0,876; 0,941) and indicated a very high agreement, with the linear correlation coefficient indicating a strong positive correlation (p <0.001). The ICC for the VHIR variable was not significant, although the linear correlation coefficient indicates a strong positive correlation (p <0.001).
Clinical Relevance Statment
This study reveals that there is good agreement in the comparison of two systems designed to analyze the movement demands of each professional soccer athlete in relation to the total distance covered. Level of Evidence I; Methodological Study - Investigation of a diagnostic test.
Keywords:
Data Accuracy; Soccer; Athletic Performance; Materials testing
... Colleges and universities are the bases for cultivating reserve talents and are the source of guaranteeing that talents in various fields of the country are constantly put into national construction. In recent years, as the requirements of talents for social development have become higher and higher, colleges and universities have carried out educational reforms accordingly [2]. In the background of the new cultural education concepts, breakthroughs and innovations are constantly made, and physical education, as an important part of college education, has an irreplaceable role in the process of students' physical and mental development and sound personality. ...
This paper constructs a random matrix-based teaching evaluation model for college volleyball clubs and deeply investigates the impact of a random matrix-based teaching evaluation model for college volleyball clubs on the effectiveness of college volleyball teaching. Based on random matrix theory (RMT), we analyze the data characteristics according to the single ring law. By introducing matrix Stein pairs, combined with the Laplace transform method, some concentration inequalities of random matrices are proved, and these inequalities play a very important role in the study of eigenvalues of random matrices. The random matrix model was used to analyze the changes brought by the club-based curriculum teaching to students’ physical quality, and a random matrix-based assessment model of college volleyball club teaching was proposed. The model fit test and independence test were conducted using IBM SPSS Statistics 20 software, and an online survey in the form of Questionnaire Star platform was used to map the correlation between college volleyball education and club-based teaching reform with X college physical education students as the research subjects, to provide more scientific theoretical guidance for the influence of club-based teaching mode of physical education courses on the physical quality of college students.
... Here, DL algorithms play an important role to analyze the recordings as exactly as possible. Cust et al. (2019) provide a wide overview of publications in the field of computer vision in different sports. It is noticeable that 39 of the 144 found publications can be assigned to football which makes up the largest proportion when we compare individual sports (Stensland et al., 2014;Manafifard et al., 2017;Linke et al., 2020). ...
This paper explores the role of artificial intelligence (AI) in elite sports. We approach the topic from two perspectives. Firstly, we provide a literature based overview of AI success stories in areas other than sports. We identified multiple approaches in the area of Machine Perception, Machine Learning and Modeling, Planning and Optimization as well as Interaction and Intervention, holding a potential for improving training and competition. Secondly, we discover the present status of AI use in elite sports. Therefore, in addition to another literature review, we interviewed leading sports scientist, which are closely connected to the main national service institute for elite sports in their countries. The analysis of this literature review and the interviews show that the most activity is carried out in the methodical categories of signal and image processing. However, projects in the field of modeling & planning have become increasingly popular within the last years. Based on these two perspectives, we extract deficits, issues and opportunities and summarize them in six key challenges faced by the sports analytics community. These challenges include data collection, controllability of an AI by the practitioners and explainability of AI results.
... In Cricket fast bowling, McNamara, Gabbet, Chapman, Naughton, & Farhart (2015) were able to use measures from the accelerometer and gyroscope to differentiate between bowling and nonbowling actions. In addition, the application machine learning in a sport context is increasing able to identify specific movements using data derived from IMUs (Crust, Sweeting, Ball, & Robertson, 2018). While such a machine learning approach has been used in badminton (Anand, Sharma, Srivastava, Kaligounder, & Prakash, 2017), this was from using two wrist worn IMUs to identify stroke type (serve, clear, drop or smash). ...
While the use of accelerometer derived Player Load has become increasingly prominent, the limitation of this approach is that training load is reduced to a single number with no differentiation between the mechanisms of loading, resulting in a loss of context. As recovery from different loadings occur at different rates, the inability to differentiate between the loading mechanisms could lead to under or over training in one or more of these mechanisms. This study sought to compare axis specific accelerometer derived Player Load with differential RPE scores to establish a means of quantifying the lower limb biomechanical load of adolescent badminton training, to try and understand some of the context into the Player Load number. It was postulated that the Player Load from the vertical axis would provide a more precise measure of lower limb loading as other loading parameters, such as upper body rotation observed during a smash, would be removed from the calculation. Nineteen adolescent badminton players (Age: 14.0 ± 0.8 y) based at a dedicated high performance youth training environment wore a GPS-embedded accelerometer between the scapulae in a purpose built vest during court-based training. After each training session the participants provided two RPE scores, one localised for the legs and one for breathlessness. Overall low correlations were observed between the Player Load and RPE values. The Player Load for the vertical axis showed a stronger correlation with the RPE for breathlessness than the RPE for the lower limb stress. The results from this study suggest that axis specific Player Load from the vertical axis does not provide greater insight into lower-limb biomechanical load compared to overall Player Load in adolescent badminton players.
... Deep neural network structures have advantages in processing time-series sensor data and require lower computational cost than traditional machine learning approaches. Integrated with deep learning algorithms, wearable sensors can be utilized for pattern recognition (Cust et al., 2019) and biomechanical variable prediction (Stetter et al., 2019;Hernandez et al., 2021) without experiment and environmental limitations. Hu et al. (2018) found that long short-term memory (LSTM) recurrent neural networks (RNN) can detect surfaceand age-related differences in walking gait based on a single wearable IMU sensor. ...
With the emergence of wearable technology and machine learning approaches, gait monitoring in real-time is attracting interest from the sports biomechanics community. This study presents a systematic review of machine learning approaches in running biomechanics using wearable sensors. Electronic databases were retrieved in PubMed, Web of Science, SPORTDiscus, Scopus, IEEE Xplore, and ScienceDirect. A total of 4,068 articles were identified via electronic databases. Twenty-four articles that met the eligibility criteria after article screening were included in this systematic review. The range of quality scores of the included studies is from 0.78 to 1.00, with 40% of articles recruiting participant numbers between 20 and 50. The number of inertial measurement unit (IMU) placed on the lower limbs varied from 1 to 5, mainly in the pelvis, thigh, distal tibia, and foot. Deep learning algorithms occupied 57% of total machine learning approaches. Convolutional neural networks (CNN) were the most frequently used deep learning algorithm. However, the validation process for machine learning models was lacking in some studies and should be given more attention in future research. The deep learning model combining multiple CNN and recurrent neural networks (RNN) was observed to extract different running features from the wearable sensors and presents a growing trend in running biomechanics.
... In this way, recent research efforts have focused on the creation of datasets by using different sensory systems. Video-based and portable technologies are the main alternatives in the monitoring of human activities 12 , so databases obtained with both systems are of great interest. However, video-based technologies entail occlusions and patients' privacy concerns 13,14 . ...
This document introduces the PHYTMO database, which contains data from physical therapies recorded with inertial sensors, including information from an optical reference system. PHYTMO includes the recording of 30 volunteers, aged between 20 and 70 years old. A total amount of 6 exercises and 3 gait variations were recorded. The volunteers performed two series with a minimum of 8 repetitions in each one. PHYTMO includes magneto-inertial data, together with a highly accurate location and orientation in the 3D space provided by the optical system. The files were stored in CSV format to ensure its usability. The aim of this dataset is the availability of data for two main purposes: the analysis of techniques for the identification and evaluation of exercises using inertial sensors and the validation of inertial sensor-based algorithms for human motion monitoring. Furthermore, the database stores enough data to apply Machine Learning-based algorithms. The participants’ age range is large enough to establish age-based metrics for the exercises evaluation or the study of differences in motions between different groups.
... Networks. FCN uses and extends existing sports performance classification models and then fine-tunes the parameters of the convolutional layers with the input of the whole image and the gold standard [15,26,27]. e convolutional layers in FCN are translation invariant; that is, the operations of the convolutional layers are only associated with low-level feature maps in the perceptual region [28,29]. ...
This paper proposes a network model recurrent fully connected network (RFC-Net) based on recurrent full convolution and polarization change. RFC-Net enriches the network by reconstructing and fine-tuning the fully convolutional network and adding recurrent convolutions to it. By studying the data mining technology of multidimensional association rules, based on the existing algorithms, this paper improves the shortcomings of the algorithms and realizes an efficient and practical method for data mining based on interdimensional multidimensional association rules. On the basis of mastering the actual student information, the effectiveness of the method is tested, and an employment analysis system based on association rules is established. Aiming at the fact that traditional grade prediction methods ignore the different influences of different behavioral characteristics on grades, and considering that behavioral data in different periods have different influences on student grades, the grade prediction problem is abstracted into a time series classification problem. The mechanism is combined with long short-term memory neural network to construct a performance prediction model based on Attention-BiLSTM. Experiments show that the prediction model proposed in this paper improves the accuracy and effectively improves the prediction quality compared with the logistic regression model with a better prediction effect in the traditional benchmark model and the long short-term memory neural network model without the introduction of the attention mechanism. Research shows that physical performance and academic performance are not contradictory. We must face up to the status of physical exercise in schools; as long as physical exercise is properly arranged, it can inspire students to form a spirit of unity, interaction, positivity, and perseverance in cultural studies.
... However, to apply bones to motion rigs, you do not need so many points. Therefore, it is one of the keys to extract joint bones from curvilinear bones by downsampling the curvilinear bones and extracting a smaller number of joint point sets suitable for bone binding and motion data loading [2]. Downsampling is a multirate digital signal processing technique or the process of reducing the sampling rate of a signal, usually used to reduce the data transmission rate or data size. ...
In order to improve the effect of tennis intelligent teaching, this paper combines the human skeleton node model to construct a tennis intelligent teaching model. Aiming at the fact that there are errors in model identification and positional disturbances such as friction which cannot be avoided in actual tennis teaching, an effective sliding mode control strategy based on disturbance observer is presented. Moreover, this paper builds a simulation model through MATLAB/Simulink to verify the influence of the three parameters of inertia, damping, and stiffness in the admittance model on the control performance. Then, this paper combines the admittance control theory to analyze the difference between the active mode and the passive mode and the main functions of the two control modes and constructs an intelligent simulation system. The experimental study shows that the tennis intelligent teaching model based on the human skeleton node model has good effects in tennis action correction and tennis-teaching effect improvement.
... Action classification is the basis of action evaluation, and, after determining the action category, it is possible to design evaluation algorithms to achieve a quantitative evaluation of human actions based on action characteristics. e study of fine classification and evaluation of human movement based on visual data can help solve three major problems in human movement understanding: distinguishing movement categories, locating movement occurrence time, and assessing movement completion quality [2]. e amount of computation decreased by 40.8%, but the accuracy increased by 0.7%. ...
This paper uses data analysis and action recognition algorithms to conduct in-depth research and analysis of professional sports competition judging and designs a professional sports competition-assisted judging system for use in actual judging. In this paper, a wearable motion capture system based on an inertial sensing unit is developed and designed for kayaking technical motion monitoring to achieve the acquisition, analysis, and quantitative evaluation of kayaker motion data. To limit the gyroscope and pose estimation error, a gradient descent method is used for multisensor data fusion to achieve athlete pose update, and a quaternion-driven human skeletal vector model is proposed to reconstruct the kayaker’s paddling technical movements. By calculating the angular sequences of the left shoulder, right shoulder, left elbow, and right elbow joints of the athlete’s upper limbs and comparing them with the optical motion capture system, the results show that the motion capture system developed in this paper is comparable to the optical motion capture system in terms of measurement accuracy. It ultimately affects the result of pose estimation. Therefore, high-resolution networks and low-resolution networks can continuously maintain high-resolution features of the image by allowing each representation layer to repeatedly accept the representation information of other networks. A step matrix model is constructed to encode the multiscale global temporal information of action sequences, and action classification is achieved by calculating the response of the step matrix of test samples to the step matrix of each category of actions. The algorithm achieves 78.96%, 91.84%, and 91.18% accuracy of action classification on the Northwestern-UCLA database, MSRC-12 database, and CAD-60 database, respectively. The designed visual motion tracking system was applied to record the motion data of the experimental subjects in the fine motion assessment task and construct the motion assessment database. The experimental results show that the average error between the prediction results of the proposed action assessment method and the manual scoring is 1.83, and the automated assessment of fine movements is effectively realized.
... Analysis of the obtained data found that, in all the impact areas, the greater the grip strength on the grip, the greater the vibration transmitted to the arm, the grip strength is reduced, and the vibration load on the arm is reduced. Literature [10] uses regression analysis to explain the relationship between longitudinal collision position and ball speed, vibration, and arm vibration. e subjects of the experiment were 19 professional athletes. ...
In order to explore the optimal control method of tennis racket string diameter, this paper applies the Kalman filter algorithm to the collection and processing of tennis sports parameters. When the Kalman filter uses the optimal gain (the Kalman filter enters the steady state), the corresponding cost function is established based on the noncorrelated nature of its residual sequence, which is used as the judgment condition for the sampling strategy switch to improve the stability of the system. In addition, this paper improves the real-time performance and calculation accuracy of data transformation through adaptive sampling strategy and adaptive scale factor, improves the stability and estimation accuracy of the system as a whole, and builds an intelligent monitoring system. Finally, this paper systematically studied the optimization control method of tennis racket string diameter and verified that the Kalman filter algorithm can play a certain role in the optimization control of tennis racket string diameter.
... Sport-specific movement recognition -Sport-specific movement recognition can be utilized for the objective performance analysis of an (elite) athlete. In that regard, Cust et al. [46] explore the automated recognition and characterization of movements in sports, which can provide an alternative for an otherwise manual, timeconsuming, limited performance analysis. The authors perform a systematical literature analysis on machine learning-and deep learning-based approaches for movement recognition in sports depending on input data from computer vision and inertial measurement units. ...
Deep learning has remarkably impacted several different scientific disciplines over the last few years. For example, in image processing and analysis, deep learning algorithms were able to outperform other cutting-edge methods. Additionally, deep learning has delivered state-of-the-art results in tasks like autonomous driving, outclassing previous attempts. There are even instances where deep learning outperformed humans, for example with object recognition and gaming. Deep learning is also showing vast potential in the medical domain. With the collection of large quantities of patient records and data, and a trend towards personalized treatments, there is a great need for automated and reliable processing and analysis of health information. Patient data is not only collected in clinical centers, like hospitals and private practices, but also by mobile healthcare apps or online websites. The abundance of collected patient data and the recent growth in the deep learning field has resulted in a large increase in research efforts. In Q2/2020, the search engine PubMed returned already over 11,000 results for the search term ‘deep learning’, and around 90% of these publications are from the last three years. However, even though PubMed represents the largest search engine in the medical field, it does not cover all medical-related publications. Hence, a complete overview of the field of ‘medical deep learning’ is almost impossible to obtain and acquiring a full overview of medical sub-fields is becoming increasingly more difficult. Nevertheless, several review and survey articles about medical deep learning have been published within the last few years. They focus, in general, on specific medical scenarios, like the analysis of medical images containing specific pathologies. With these surveys as a foundation, the aim of this article is to provide the first high-level, systematic meta-review of medical deep learning surveys.
... This is a lighter, more ecological, and easier to use alternative method for the measurement of multiple times over long distances in alpine skiing and other sports, regardless of the place where they are practiced. In addition, the use of IMUs in sports training has made it easier to obtain kinematic and kinetic data on movement [26][27][28][29][30][31]. The fact that all the built-in sensors (accelerometer, gyroscope, GNSS receiver, magnetometer, etc.) are synchronized optimizes data collection. ...
Citation: Buxadé, C.P.-C.; Fernández-Valdés, B.; Morral-Yepes, M.; Viñas, S.T.; Riu, J.M.P.; Moras Feliu, G. Validity of a Magnet-Based Timing System Using the Magnetometer Built into an IMU. Abstract: Inertial measurement units (IMUs) represent a technology that is booming in sports right now. The aim of this study was to evaluate the validity of a new application on the use of these wearable sensors, specifically to evaluate a magnet-based timing system (M-BTS) for timing short-duration sports actions using the magnetometer built into an IMU in different sporting contexts. Forty-eight athletes (22.7 ± 3.3 years, 72.2 ± 10.3 kg, 176.9 ± 8.5 cm) and eight skiers (17.4 ± 0.8 years, 176.4 ± 4.9 cm, 67.7 ± 2.0 kg) performed a 60-m linear sprint running test and a ski slalom, respectively. The M-BTS consisted of placing several magnets along the course in both contexts. The magnetometer built into the IMU detected the peak-shaped magnetic field when passing near the magnets at a certain speed. The time between peaks was calculated. The system was validated with photocells. The 95% error intervals for the total times were less than 0.077 s for the running test and 0.050 s for the ski slalom. With the M-BTS, future studies could select and cut the signals belonging to the other sensors that are integrated in the IMU, such as the accelerometer and the gyroscope.
... To recognize each video frame in an action, the model must be trained individually for that frame, and the parameters must be changed regularly to reflect data changes. This will result in a higher computational cost and a significant potential for catastrophic forgetfulness [18,26,27]. ...
With the fast development of sports in recent years, the number of people participating in various sports has increased day by day. Because of the advantages of fewer constraints on the field and ease of learning, badminton became one of the most popular sports among them. Numerous works have been done specifically for the recognition of action of badminton players to improve and popularize it, but the traditional badminton player’s badminton action recognition algorithm employs the method of manually constructing a topology map to model the action sequence contained in multiple video frames. Besides, it learns each video frame in a targeted manner to reflect data changes, which is prone to high computational cost, low network generalization, and catastrophic forgetting. In response to the above problems, this paper proposes a deep learning-based action recognition technology for badminton players, which re-encodes the human hitting action sequence data with multirelational characteristics into relational triples, and learns by decoupling based on long short-term memory network. At the same time, this paper designs and completes a set of badminton action recognition schemes based on acceleration and angular velocity signals. Experimental results show that the proposed method achieves 63%, 84%, and 92%, respectively, recognition accuracy on multiple benchmark datasets, which improves the accuracy of human hitting action recognition. As a result, the evaluation will be useful in future work to improve the structure of current deep learning models for higher results in badminton action recognition.
... The use of IMUs is well established in the literature [16]. There is support for the use of IMUs as tracking devices for human movement in sport and translating specific movement actions which can be detected and extracted using IMUs [17]. A recent study suggested that the use of microprocessor technology (such as that contained in IMUs) is a valid method to detect human motion in contact events in rugby union [18]. ...
Rugby union is a field sport that is played at amateur and professional levels by male and female players globally. One of the most prevalent injury risks associated with the sport involves tackle collisions with opposition players. This suggests that a targeted injury reduction strategy could focus on the tackle area in the game. In amateur rugby union, injuries to the head, face and shoulder are the most common injury sites in youth rugby playing populations. A suboptimal tackle technique may contribute to an increased injury risk in these populations. One proposed mitigation strategy to reduce tackle-related injuries in youth populations may be to increase tackle proficiency by coaching an effective tackle technique. The present study aimed to demonstrate a proof of concept for a tackle technique coaching platform using inertial measurement units (IMUs) and a bespoke mobile application developed for a mobile device (i.e., a mobile phone). The test battery provided a proof of concept for the primary objective of modelling the motion of a player in a tackle event. The prototype (bespoke mobile application) modelled the IMU in a 3D space and demonstrated the orientation during a tackle event. The participants simulated ten tackle events that were ten degrees above and ten degrees below the zero degree of approach, and these (unsafe tackles) were indicated by a red light on the mobile display unit. The parameters of ten degrees above and below the zero angle of approach were measured using an inclinometer mobile application. These tackle event simulations provided a real-time stream of data that displayed the angle of tackles on a mobile device. The novel coaching platform could therefore constitute part of an injury reduction strategy for amateur or novice coaches to instruct safer tackle practice in youth rugby playing populations.
... Performing real-time predictions based on machine learning and providing realtime feedback has a huge potential to both enhance the athlete's performance, through movement recognition [264] and technique correction [23], and to assess injury predictive factors [265] and evaluate them within long-term monitoring of injury-forecasting systems [240]. This is particularly relevant for athletes with disabilities, where personalised monitoring systems can be developed to anticipate injuries specific to their disability [26]. ...
Wearable technologies are often indicated as tools that can enable the in-field collection of quantitative biomechanical data, unobtrusively, for extended periods of time, and with few spatial limitations. Despite many claims about their potential for impact in the area of injury prevention and management, there seems to be little attention to grounding this potential in biomechanical research linking quantities from wearables to musculoskeletal injuries, and to assessing the readiness of these biomechanical approaches for being implemented in real practice. We performed a systematic scoping review to characterise and critically analyse the state of the art of research using wearable technologies to study musculoskeletal injuries in sport from a biomechanical perspective. A total of 4952 articles were retrieved from the Web of Science, Scopus, and PubMed databases; 165 were included. Multiple study features—such as research design, scope, experimental settings, and applied context—were summarised and assessed. We also proposed an injury-research readiness classification tool to gauge the maturity of biomechanical approaches using wearables. Five main conclusions emerged from this review, which we used as a springboard to propose guidelines and good practices for future research and dissemination in the field.
... Along these lines, some authors have succeeded in using various learning machines to classify a situation or recognise movements of objects or people based on data provided by movement sensors from smart mobile devices (Rodriguez-Martin et al., 2013;Tian et al., 2019;Cust et al., 2019). In this paper we have selected four supervised machine learning classification techniques: linear discriminant analysis (LDA) used to classify linearly separable samples, multilayer perceptron (MLP) and support vector machine (SVM) used to classify non-linearly separable samples and Bayesian classifier or probabilistic neural network (PNN). ...
Knowing the activity of fishing vessels accurately and in real time means a leap in quality in the management of fishing activity. This paper presents the development of a new fishing activity monitoring integral system (FAMIS) that can complement and overcome the limitations of current fishing vessel monitoring systems (VMS). FAMIS is developed on the basis of a low-cost mobile device with GPS sensors, accelerometer, gyroscope and magnetic field and integrates different statistical methods (discriminant functions) and heuristics (artificial neural networks and vectorial support machines) as techniques to classify the information recorded by the sensors of a mobile device during fishing activity. The results obtained with FAMIS indicate that, in general, heuristics have a high degree of discrimination of each of the phases of fishing operation and that, in particular, multilayer perceptrons (MLPs) are capable of correctly identifying 96.3% of towing phases using only GPS and gyro sensors.
... A filtering method is also proposed in the pedometry algorithm study of [15], which normalises the peak features so that an acceleration point with a value of 1 after filtering represents the generation of a step. In the work on the pedometry algorithm of [16], the authors extracted features such as the number of relationships and the degree of deviation from positive and negative values of the acceleration data, and then set thresholds for each feature; only acceleration peaks that reached the threshold setting were selected as candidate peaks. In the 3D positioning system research work of [17], the authors implemented an adaptive peak counting algorithm that improves the adaptability of the peak detection pedometry algorithm under multiple movements by setting different peak thresholds for different movements. ...
To investigate the effectiveness of pedometry in athletics at different running speeds and on different walking surfaces, and to test whether it can be used for running for fitness and for pedometry on hardened concrete surfaces in daily life. The steps were measured under laboratory conditions using a running platform at 5 min at 5 speeds of 3.2 km/h, 4.8 km/h, 6.4 km/h, 8.0 km/h, and 9.6 km/h (actual steps were accurately determined by video playback), followed by 300 steps each at low, normal and fast speeds on an outdoor plastic athletics track. There were significant differences between the smart bracelet and the mobile phone sports app at low running speeds on four different surfaces: outdoor plastic track, dirt, concrete and mountainous terrain. The difference was not significant in normal pace and fast walking.
... Also, measurements related to the movements and actions performed by the athlete help to estimate the external WL. Automatic activity assessment is possible via several tools such as video monitoring [2,3] or signal processing algorithms [4]. However, video and radio-based local positioning systems require equipping the gym, which is restrictive. ...
Movement data from athletes are useful to quantify performance or more specifically the workload. Inertial measurement units (IMUs) are useful sensors to quantify body movements. Sensor placement on human body is still an open question that this paper focuses on. A method that develops synthesized inertial data is proposed for determining optimal sensors placement. Comparison between virtual and real inertial data is achieved. Training motion recognition algorithm on synthesized and real inertial data exhibits less than 7% difference. This method highlights the ability of the numerical model to determine relevant sensor placement of IMUs on human body for motion recognition algorithm using virtual sensors.
... Detection of playfield region has two objectives. One is to detect the playfield region from non-playfield areas as presented in [7], while the other is to identify primary objects from the background by filtering out redundant pixels such as grass, court lines. This provides a reduced pixel which requires processing and reduction of errors for simplifying player or ball detection and tracking phases, event extraction, pose detection etc. ...
Recent developments in video analysis of sports and computer vision techniques have achieved significant improvements to enable a variety of critical operations. To provide enhanced information, such as detailed complex analysis in sports like soccer, basketball, cricket, badminton, etc., studies have focused mainly on computer vision techniques employed to carry out different tasks. This paper presents a comprehensive review of sports video analysis for various applications high-level analysis such as detection and classification of players, tracking player or ball in sports and predicting the trajectories of player or ball, recognizing the teams strategies, classifying various events in sports. The paper further discusses published works in a variety of application-specific tasks related to sports and the present researchers views regarding them. Since there is a wide research scope in sports for deploying computer vision techniques in various sports, some of the publicly available datasets related to a particular sport have been provided. This work reviews a detailed discussion on some of the artificial intelligence(AI)applications in sports vision, GPU-based work stations, and embedded platforms. Finally, this review identifies the research directions, probable challenges, and future trends in the area of visual recognition in sports.
... Recently, researchers have attracted significant attention to Deep Learning (DL) [11][12][13][14] owing to its numerous applications in speech processing [15], natural language processing [16], and CV [17,18]. In video recognition [19] and large-scale images, a model of DL so-called convolutional neural network (CNN) has lately attained several encouraging results. Simonyan and Zisserman [20] introduced in 2015 the VGG model, which accomplished an extremely well-behaved performance with ImageNet and became used-widely in different studies of CV. ...
In the last decade, there has been a surge of interest in addressing complex Computer Vision (CV) problems in the field of face recognition (FR). In particular, one of the most difficult ones is based on the accurate determination of the ethnicity of mankind. In this regard, a new clas‐ sification method using Machine Learning (ML) tools is proposed in this paper. Specifically, a new Deep Learning (DL) approach based on a Deep Convolutional Neural Network (DCNN) model is developed, which outperforms a reliable determination of the ethnicity of people based on their facial features. However, it is necessary to make use of specialized high‐performance computing (HPC) hardware to build a workable DCNN‐based FR system due to the low computation power given by the current central processing units (CPUs). Recently, the latter approach has increased the efficiency of the network in terms of power usage and execution time. Then, the usage of field‐ programmable gate arrays (FPGAs) was considered in this work. The performance of the new DCNN‐based FR method using FPGA was compared against that using graphics processing units (GPUs). The experimental results considered an image dataset composed of 3141 photographs of citizens from three distinct countries. To our knowledge, this is the first image collection gathered specifically to address the ethnicity identification problem. Additionally, the ethnicity dataset was made publicly available as a novel contribution to this work. Finally, the experimental results proved the high performance provided by the proposed DCNN model using FPGAs, achieving an accuracy level of 96.9 percent and an F1 score of 94.6 percent while using a reasonable amount of energy and hardware resources.
Artificial intelligence and its subcategories of machine learning and deep learning are gaining increasing importance and attention in the context of sports research. This has also meant that the number of corresponding publications has become complex and unmanageably large in human terms. In the current state of the research field, there is a lack of bibliometric analysis, which would prove useful for obtaining insights into the large amounts of available literature. Therefore, the present work aims to identify important research issues, elucidate the conceptual structure of the research field, and unpack the evolutionary trends and the direction of hot topics regarding key themes in the research field of artificial intelligence in sports. Using the Scopus database, 1215 documents (reviews and articles) were selected. Bibliometric analysis was performed using VOSviewer and bibliometrix R package. The main findings are as follows: (a) the literature and research interest concerning AI and its subcategories is growing exponentially; (b) the top 20 most cited works comprise 32.52% of the total citations; (c) the top 10 journals are responsible for 28.64% of all published documents; (d) strong collaborative relationships are present, along with small, isolated collaboration networks of individual institutions; (e) the three most productive countries are China, the USA, and Germany; (f) different research themes can be characterized using author keywords with current trend topics, e.g., in the fields of biomechanics, injury prevention or prediction, new algorithms, and learning approaches. AI research activities in the fields of sports pedagogy, sports sociology, and sports economics seem to have played a subordinate role thus far. Overall, the findings of this study expand knowledge on the research situation as well as the development of research topics regarding the use of artificial intelligence in sports, and may guide researchers to identify currently relevant topics and gaps in the research.
The use of wearable devices for player motion analysis is becoming a popular method to measure kinematic parameters associated with player techniques in racket sports. This systematic review focused on identifying the current applications of wearable technology for player motion analysis in racket sports (tennis, table tennis, badminton and squash) through two research questions: (1) What are the existing applications of wearable technology for player motion analysis in racket sports? (2) What data analysis methods are used to quantify and evaluate player motion? A comprehensive search of MEDLINE, EMBASE, SPORTDiscus, Scopus, Web of Science, and IEEE databases was undertaken following PRISMA reporting guidelines. Included studies must have only used external wearable technology mounted to either the player or the racket for the potential application of player motion analysis in a racket sport, tested the wearable technology under normal playing conditions and not only focused on detecting/classifying the player's stroke or activity movement. Of the 6616 articles found, a total of 15 studies met the inclusion and exclusion criteria. Tennis and table tennis were the most popular sports researched and inertial measurement units and electromyography sensors were the most common types of sensors used. This review found that wearable devices were mainly used to analyze (1) movement similarities and differences of players at different playing levels, (2) variability in racket, upper limb and joint movement patterns and (3) movement differences associated with different ball spin levels. These technologies can create a new level of accessible personalized training in these sports.
Emotion is an important research topic in the field of sports. The physiological changes caused by emotion have a great influence on the completion of sports. It cannot only fully mobilize the organism and maximize the exercise potential, but also lead to muscle stiffness, movement deformation, or muscle contraction weakness. Furthermore, it can affect the completion of exercise. In order to ensure the athlete can keep the best competitive level, it is necessary to estimate the athlete's emotion before competition. This paper adopts the pulse wave signal to implement the emotion estimation for the athletes. First, the pulse wave signals are collected by using a portable sensor via mobile computing. Then, the collected pulse wave signals have noises removed by wavelet transform. Last, the denoised pulse wave signals are represented as the features in time domain and frequency domain to input into a trained classifier for determining the current emotion status. The experimental results show that the proposed method can recognize more than 90% of the abnormal emotions.
To understand human behaviors, action recognition based on videos is a common approach. Compared with image-based action recognition, videos provide much more information. Reducing the ambiguity of actions and in the last decade, many works focused on datasets, novel models and learning approaches have improved video action recognition to a higher level. However, there are challenges and unsolved problems, in particular in sports analytics where data collection and labeling are more sophisticated, requiring sport professionals to annotate data. In addition, the actions could be extremely fast and it becomes difficult to recognize them. Moreover, in team sports like football and basketball, one action could involve multiple players, and to correctly recognize them, we need to analyse all players, which is relatively complicated. In this paper, we present a survey on video action recognition for sports analytics. We introduce more than ten types of sports, including team sports, such as football, basketball, volleyball, hockey and individual sports, such as figure skating, gymnastics, table tennis, tennis, diving and badminton. Then we compare numerous existing frameworks for sports analysis to present status quo of video action recognition in both team sports and individual sports. Finally, we discuss the challenges and unsolved problems in this area and to facilitate sports analytics, we develop a toolbox using PaddlePaddle, which supports football, basketball, table tennis and figure skating action recognition.
Purpose:
Maintaining balance during static standing postures requires the coordination of many neuromuscular mechanisms. The role of the intrinsic and extrinsic foot muscles in this paradigm has yet to be clearly defined. The purpose of this study was to explore foot muscle activation during static phases on common weight-bearing tasks of varying loads and balance demands.
Methods:
Twenty healthy young adults performed 6 standing postures (single-limb and double-limb stand, squat, and heel raise) with one foot on a force plate. Muscle activity was recorded from the abductor hallucis, flexor hallucis longus and brevis, and tibialis posterior using intramuscular electrodes; surface electrodes were used to record activity from the peroneus longus and tibialis anterior. Two-way repeated measures ANOVA (2 loading conditions × 3 postures) were run to compare muscle activation and center of pressure velocity.
Results:
Intrinsic foot muscle activity increased as loading and postural demand increased; however, the specific effects varied for each of the extrinsic foot muscles.
Conclusions:
These results suggest that the intrinsic foot muscles play an important role in maintaining static balance. Strengthening intrinsic and extrinsic foot muscles may help increase stability in people who have weak toe flexors or who suffer from a variety of foot pathologies.
Home-based physical therapies are effective if the prescribed exercises are correctly executed and patients adhere to these routines. This is specially important for older adults who can easily forget the guidelines from therapists. Inertial Measurement Units (IMUs) are commonly used for tracking exercise execution giving information of patients’ motion data. In this work, we propose the use of Machine Learning techniques to recognize which exercise is being carried out and to assess if the recognized exercise is properly executed by using data from four IMUs placed on the person limbs. To the best of our knowledge, both tasks have never been addressed together as a unique complex task before. However, their combination is needed for the complete characterization of the performance of physical therapies. We evaluate the performance of six machine learning classifiers in three contexts: recognition and evaluation in a single classifier, recognition of correct exercises, excluding the wrongly performed exercises, and a two-stage approach that first recognizes the exercise and then evaluates it. We apply our proposal to a set of 8 exercises of the upper-and lower-limbs designed for maintaining elderly people health status. To do so, the motion of 30 volunteers were monitored with 4 IMUs. We obtain accuracies of 88.4 % and the 91.4 % in the two initial scenarios. In the third one, the recognition provides an accuracy of 96.2 %, whereas the exercise evaluation varies between 93.6 % and 100.0 %. This work proves the feasibility of IMUs for a complete monitoring of physical therapies in which we can get information of which exercise is being performed and its quality, as a basis for designing virtual coaches.
Cette thèse avait pour objectif de présenter les différents travaux réalisés sur la prédiction de la performance en course à pied afin d’aider les athlètes et les entraîneurs à optimiser leur processus d’entraînement. Ces études, en collaboration avec la Fédération Française d’Athlétisme (FFA), se sont appuyées sur le système d’information fédéral répertoriant notamment l’ensemble des résultats athlétiques, les bilans ou encore le nombre de licenciés. La première étude avait pour objectif d’exposer l’évolution des performances françaises des courses de demi-fond et de fond chez les femmes. Les études suivantes étaient principalement destinées à tester la validité, la justesse, et la précision de différentes méthodes de prédiction (i.e., capacité à prédire les performances) sur des performances individuelles réelles d’athlètes de différents niveaux, hommes et/ou femmes. Les résultats se sont avérés valides et précis, quelle que soit la méthode de prédiction utilisée. Enfin, la dernière recherche était destinée à la prédiction du potentiel de performance. Cette étude a notamment mis en avant une analyse du taux d'amélioration des performances en demi-fond et en fond précédant la réalisation de records personnels chez les hommes et chez les femmes. Un index de progression à visée pratique, a également été proposé, afin d’évaluer l’évolution des performances et permettre une éventuelle détection et orientation des athlètes au fort potentiel.
In recent years smart sport equipments have achieved great success in professional and amateur sports, as well as body sensory systems; now discovering interesting knowledge in the surge of data from those embedded sensors used in sports is necessary and the focus of our research. In this paper, we investigate golf swing data classification method based on deep convolutional neural network (deep CNN) fed with multi-sensor golf swing signals. Our smart golf club integrates two orthogonally affixed strain gage sensors, 3-axis accelerometer and 3-axis gyroscope, and collects real-world golf swing data from professional and amateur golf players. Furthermore we explore the performance of our well-trained CNN-based classifier and evaluate it on the real-world test set in terms of common indicators including accuracy, precision-recall, and F1-score. Experiments and corresponding results show that our CNN-based model can satisfy the requirement of accuracy of golf swing classification, and outperforms support vector machine (SVM) method.
Recent technological developments have led to the production of inexpensive, non-invasive, miniature magneto-inertial sensors, ideal for obtaining sport performance measures during training or competition. This systematic review evaluates current evidence and the future potential of their use in sport performance evaluation. Articles published in English (April 2017) were searched in Web-of-Science, Scopus, Pubmed, and Sport-Discus databases. A keyword search of titles, abstracts and keywords which included studies using accelerometers, gyroscopes and/or magnetometers to analyse sport motor-tasks performed by athletes (excluding risk of injury, physical activity, and energy expenditure) resulted in 2040 papers. Papers and reference list screening led to the selection of 286 studies and 23 reviews. Information on sport, motor-tasks, participants, device characteristics, sensor position and fixing, experimental setting and performance indicators was extracted. The selected papers dealt with motor capacity assessment (51 papers), technique analysis (163), activity classification (19), and physical demands assessment (61). Focus was placed mainly on elite and sub-elite athletes (59%) performing their sport in-field during training (62%) and competition (7%). Measuring movement outdoors created opportunities in winter sports (8%), water sports (16%), team sports (25%), and other outdoor activities (27%). Indications on the reliability of sensor-based performance indicators are provided, together with critical considerations and future trends.
Background:
Analysis of lower limb exercises is traditionally completed with four distinct methods (i) 3D motion capture; (ii) depth-camera based systems (iii) visual analysis from a qualified exercise professional; (iv) self-assessment. Each method is associated with a number of limitations.
Objective:
The aim of this systematic review is to synthesize and evaluate studies which have investigated the capacity for inertial measurement unit (IMU) technologies to assess movement quality in lower limb exercises.
Data Sources:
A systematic review of PubMed, ScienceDirect and Scopus was conducted.
Study Eligibility Criteria:
Articles written in English and published in the last 10 years which contained an IMU system for the analysis of repetition-based targeted lower limb exercises were included.
Study Appraisal and Synthesis Methods:
The quality of included studies was measured using an adapted version of the STROBE assessment criteria for cross-sectional studies. The studies were categorised in to three groupings: exercise detection, movement classification or measurement validation. Each study was then qualitatively summarised.
Results:
From the 2452 articles that were identified with the search strategies, 47 papers are included in this review.
Conclusions:
Wearable inertial sensor systems for analysing lower limb exercises are a rapidly growing technology. Research over the past ten years has predominantly focused on validating measurements that the systems produce and classifying users’ exercise quality. There have been very few user evaluation studies and no clinical trials in this field to date.
The automatic classification of sub-techniques in classical cross-country skiing provides unique possibilities for analyzing the biomechanical aspects of outdoor skiing. This is currently possible due to the miniaturization and flexibility of wearable inertial measurement units (IMUs) that allow researchers to bring the laboratory to the field. In this study, we aimed to optimize the accuracy of the automatic classification of classical cross-country skiing sub-techniques by using two IMUs attached to the skier’s arm and chest together with a machine learning algorithm. The novelty of our approach is the reliable detection of individual cycles using a gyroscope on the skier’s arm, while a neural network machine learning algorithm robustly classifies each cycle to a sub-technique using sensor data from an accelerometer on the chest. In this study, 24 datasets from 10 different participants were separated into the categories training-, validation- and test-data. Overall, we achieved a classification accuracy of 93.9% on the test-data. Furthermore, we illustrate how an accurate classification of sub-techniques can be combined with data from standard sports equipment including position, altitude, speed and heart rate measuring systems. Combining this information has the potential to provide novel insight into physiological and biomechanical aspects valuable to coaches, athletes and researchers.
Advanced machine learning technologies are seldom applied to wearable motion sensor data obtained from sport movements. In this work, we therefore investigated neural networks for motion performance evaluation utilizing a set of inertial sensor-based ski jump measurements. A multi-dimensional convolutional network model that related the motion data under aspects of time, placement and sensor type was implemented. Additionally, its applicability as a measure for automatic motion style judging was evaluated. Results indicate that one multi-dimensional convolutional layer is sufficient to recognize relevant performance error representations. Furthermore, comparisons against a Support Vector Machine and a Hidden Markov Model show that the new model out-performs feature-based methods under noisy and biased data environments. Architectures such as the proposed evaluation system can hence become essential for automatic performance analysis and style judging systems in future.
Analysts in professional team sport regularly perform analysis to gain strategic and tactical insights into player and team behavior. Goals of team sport analysis regularly include identification of weaknesses of opposing teams, or assessing performance and improvement potential of a coached team. Current analysis workflows are typically based on the analysis of team videos. Also, analysts can rely on techniques from Information Visualization, to depict e.g., player or ball trajectories. However, video analysis is typically a time-consuming process, where the analyst needs to memorize and annotate scenes. In contrast, visualization typically relies on an abstract data model, often using abstract visual mappings, and is not directly linked to the observed movement context anymore. We propose a visual analytics system that tightly integrates team sport video recordings with abstract visualization of underlying trajectory data. We apply appropriate computer vision techniques to extract trajectory data from video input. Furthermore, we apply advanced trajectory and movement analysis techniques to derive relevant team sport analytic measures for region, event and player analysis in the case of soccer analysis. Our system seamlessly integrates video and visualization modalities, enabling analysts to draw on the advantages of both analysis forms. Several expert studies conducted with team sport analysts indicate the effectiveness of our integrated approach.
Human activity recognition (HAR) aims to recognize activities from a series of observations on the actions of subjects and the environmental conditions. The vision-based HAR research is the basis of many applications including video surveillance, health care, and human-computer interaction (HCI). This review highlights the advances of state-of-the-art activity recognition approaches, especially for the activity representation and classification methods. For the representation methods, we sort out a chronological research trajectory from global representations to local representations, and recent depth-based representations. For the classification methods, we conform to the categorization of template-based methods, discriminative models, and generative models and review several prevalent methods. Next, representative and available datasets are introduced. Aiming to provide an overview of those methods and a convenient way of comparing them, we classify existing literatures with a detailed taxonomy including representation and classification methods, as well as the datasets they used. Finally, we investigate the directions for future research.
Human Activity Recognition has witnessed a significant progress in the last decade. Although a great deal of work in this field goes in recognizing normal human activities, few studies focused on identifying motion in sports. Recognizing human movements in different sports has high impact on understanding the different styles of humans in the play and on improving their performance. As deep learning models proved to have good results in many classification problems, this paper will utilize deep learning to classify cross-country skiing movements, known as gears, collected using a 3D accelerometer. It will also provide a comparison between different deep learning models such as convolutional and recurrent neural networks versus standard multi-layer perceptron. Results show that deep learning is more effective and has the highest classification accuracy.
One of the hot topics in modern era of cricket is to decide whether the bowling action of a bowler is legal or not. Because of the complex bio-mechanical movement of the bowling arm, it is not possible for the on-field umpire to declare a bowling action as legal or illegal. Inertial sensors are currently being used for activity recognition in cricket for the coaching of bowlers and detecting the legality of their moves, since a well trained and legal bowling action is highly significant for the career of a cricket player. After extensive analysis and research, we present a system to detect the legality of the bowling action based on real time multidimensional physiological data obtained from the inertial sensors mounted on the bowlers arm. We propose a method to examine the movement of the bowling arm in the correct rotation order with a precise angle. The system evaluates the bowling action using various action profiles. The action profiles are used so as to simplify the complex bio-mechanical movement of the bowling arm along with minimizing the size of the data provided to the classifier. The events of interest are identified and tagged. Algorithms such as support vector machines, k-nearest neighbor, Naïve Bayes, random forest, and artificial neural network are trained over statistical features extracted from the tagged data. To accomplish the reliability of outcome measures, the technical error of measurement was adopted. The proposed method achieves very high accuracy in the correct classification of bowling action.
The deadlift is a compound full-body exercise that is fundamental in resistance training, rehabilitation programs and powerlifting competitions. Accurate quantification of deadlift biomechanics is important to reduce the risk of injury and ensure training and rehabilitation goals are achieved. This study sought to develop and evaluate deadlift exercise technique classification systems utilising Inertial Measurement Units (IMUs), recording at 51.2 Hz, worn on the lumbar spine, both thighs and both shanks. It also sought to compare classification quality when these IMUs are worn in combination and in isolation. Two datasets of IMU deadlift data were collected. Eighty participants first completed deadlifts with acceptable technique and 5 distinct, deliberately induced deviations from acceptable form. Fifty-five members of this group also completed a fatiguing protocol (3-Repition Maximum test) to enable the collection of natural deadlift deviations. For both datasets, universal and personalised random-forests classifiers were developed and evaluated. Personalised classifiers outperformed universal classifiers in accuracy, sensitivity and specificity in the binary classification of acceptable or aberrant technique and in the multi-label classification of specific deadlift deviations. Whilst recent research has favoured universal classifiers due to the reduced overhead in setting them up for new system users, this work demonstrates that such techniques may not be appropriate for classifying deadlift technique due to the poor accuracy achieved. However, personalised classifiers perform very well in assessing deadlift technique, even when using data derived from a single lumbar-worn IMU to detect specific naturally occurring technique mistakes.
The popularity of running has increased in recent years. A rise in the incidence of running-related overuse musculoskeletal injuries has occurred parallel to this. This study investigates the capability of using data from a single inertial measurement unit (IMU) to differentiate between running form in a non-fatigued and fatigued state. Data was captured from an IMU placed on the lumbar spine, right shank and left shank in 21 recreational runners (10 male, 11 female) during separate 400m running trials. The trials were performed prior to and following a fatiguing protocol. Following stride segmentation, IMU signal features were extracted from the labelled (non-fatigued vs fatigued) sensor data and used to train both a Global and Personalised classifier for each individual IMU location. A single IMU on the Lumbar spine displayed 75% accuracy, 73% sensitivity and 77% specificity when using a Global Classifier. A single IMU on the Right Shank displayed 100% accuracy, 100% sensitivity and 100% specificity when using a Personalised Classifier. These results indicate that a single IMU has the potential to differentiate between non-fatigued and fatigued running states with a high level of accuracy.
To date, the monitoring of fast bowling workloads across training and competition environments has been limited to counting total balls bowled. However, bowling at faster velocities is likely to require greater effort while also placing greater load on the bowler. This study investigated the relationship between prescribed effort and microtechnology outputs in fast bowlers to ascertain whether the technology could provide a more refined measure of workload. Twelve high performing fast bowlers (mean ± SD age; 20.3 ± 2.2 yr) participated in this study. Each bowler bowled 6 balls at prescribed bowling intensities of 60%, 70%, 85% and 100%. The relationship between microtechnology outputs, prescribed intensity and ball velocity were determined using polynomial regression. Very large relationships were observed between prescribed effort and ball velocity for peak PlayerLoad(TM) (R = 0.83 ± 0.19 and 0.82 ± 0.20). The Player Load(TM) across lower ranges of prescribed effort exhibited higher coefficient of variation (CV) [60% = 19.0 (17.0 - 23.0)%] while the CV at higher ranges of prescribed effort was lower [100% = 7.3 (6.4 - 8.5)%]. Routinely used wearable microtechnology devices offer opportunities to examine workload and intensity in cricket fast bowlers outside the normal metrics reported. They offer a useful tool for prescribing and monitoring bowling intensity and workload in elite fast bowlers.
The world of sports intrinsically involves fast and accurate motion that is not only challenging for competitors to master, but can be difficult for coaches and trainers to analyze, and for audiences to follow. The nature of most sports means that monitoring by the use of sensors or other devices fixed to players or equipment is generally not possible. This provides a rich set of opportunities for the application of computer vision techniques to help the competitors, coaches and audience. This paper discusses a selection of current commercial applications that use computer vision for sports analysis, and highlights some of the topics that are currently being addressed in the research community. A summary of on-line datasets to support research in this area is included.
Lunges are a common, compound lower-limb resistance exercise. If completed with aberrant technique, the increased stress on the joints used may increase risk of injury. This study sought to first investigate the ability of inertial measurement units (IMUs), when used in isolation and combination, to (a) classify acceptable and aberrant lunge technique (b) classify exact deviations in lunge technique. We then sought to investigate the most important features and establish the minimum number of top-ranked features and decision trees that are needed to maintain maximal system classification efficacy. Eighty volunteers performed the lunge with acceptable form and 11 deviations. Five IMUs positioned on the lumbar spine, thighs and shanks recorded these movements. Time and frequency domain features were extracted from the IMU data and used to train and test a variety of classifiers. A single IMU system achieved 83% accuracy, 62% sensitivity and 90% specificity in binary classification and a five IMU system achieved 90% accuracy, 80% sensitivity and 92% specificity. A five IMU set-up can also detect specific deviations with 70% accuracy. System efficiency was improved and classification quality was maintained when using only 20% of the top-ranked features for training and testing classifiers.
Deep neural networks (DNNs) are currently widely used for many artificial intelligence (AI) applications including computer vision, speech recognition, and robotics. While DNNs deliver state-of-the-art accuracy on many AI tasks, it comes at the cost of high computational complexity. Accordingly, techniques that enable efficient processing of deep neural network to improve energy-efficiency and throughput without sacrificing performance accuracy or increasing hardware cost are critical to enabling the wide deployment of DNNs in AI systems. This article aims to provide a comprehensive tutorial and survey about the recent advances towards the goal of enabling efficient processing of DNNs. Specifically, it will provide an overview of DNNs, discuss various platforms and architectures that support DNNs, and highlight key trends in recent efficient processing techniques that reduce the computation cost of DNNs either solely via hardware design changes or via joint hardware design and network algorithm changes. It will also summarize various development resources that can enable researchers and practitioners to quickly get started on DNN design, and highlight important benchmarking metrics and design considerations that should be used for evaluating the rapidly growing number of DNN hardware designs, optionally including algorithmic co-design, being proposed in academia and industry. The reader will take away the following concepts from this article: understand the key design considerations for DNNs; be able to evaluate different DNN hardware implementations with benchmarks and comparison metrics; understand trade-offs between various architectures and platforms; be able to evaluate the utility of various DNN design techniques for efficient processing; and understand of recent implementation trends and opportunities.
This paper presents a miniature wearable device and a system for detecting and recording the movement and biometric information of a user during sport activities. The wearable device is designed to be worn on a wrist and can monitor skin temperature and pulse rate. Furthermore it can monitor arm movement and detect gestures using inertial measurement unit. The device can be used for various professional and amateur sport applications and for health monitoring. Because of its small size and minimum weight, it is especially appropriate for swing-based sports like tennis or golf, where any additional weight on the arms would most likely disturb the player and have some influence on the player’s performance. Basic signal processing is performed directly on the wearable device but for more complex signal analysis the data can be uploaded via the Internet to a cloud service, where it can be processed by a dedicated application. The device is powered by a light-weight miniature LiPo battery and has about 6 hours of autonomy at maximum performance.
Many injuries in sports are caused by overuse. These injuries are a major cause for reduced performance of professional and non-professional beach volleyball players. Monitoring of player actions could help identifying and understanding risk factors and prevent such injuries. Currently, time-consuming video examination is the only option for detailed player monitoring in beach volleyball. The lack of a reliable automatic monitoring system impedes investigations about the risk factors of overuse injuries. In this work, we present an unobtrusive automatic monitoring system for beach volleyball based on wearable sensors. We investigate the possibilities of Deep Learning in this context by designing a Deep Convolutional Neural Network for sensor-based activity classification. The performance of this new approach is compared to five common classification algorithms. With our Deep Convolutional Neural Network, we achieve a classification accuracy of 83.2%, thereby outperforming the other classification algorithms by 16.0%. Our results show that detailed player monitoring in beach volleyball using wearable sensors is feasible. The substantial performance margin between established methods and our Deep Neural Network indicates that Deep Learning has the potential to extend the boundaries of sensor-based activity recognition.
Strength and conditioning (S&C) coaches offer expert guidance to help those they work with achieve their personal fitness goals. However, due to cost and availability issues, individuals are often left training without expert supervision. Recent developments in inertial measurement units (IMUs) and mobile computing platforms have allowed for the possibility of unobtrusive motion tracking systems and the provision of real time individualised feedback regarding exercise performance. These systems could enable S&C coaches to remotely monitor sessions and help gym users record workouts. One component of these IMU systems is the ability to identify the exercises completed. In this study, IMUs were positioned on the lumbar spine, thighs and shanks on 82 healthy participants. Participants completed 10 repetitions of the squat, lunge, single leg squat, deadlift and tuck jump with acceptable form. Descriptive features were extracted from the IMU signals for each repetition of each exercise and these were used to train an exercise classifier. The exercises were detected with 99% accuracy when using signals from all five IMUs, 98% when using signals from the thigh and lumbar IMUs and 98% with just a single IMU on the shank. These results indicate that a single IMU can accurately distinguish between five common multi-joint exercises.
Objectives:
Collision frequency during rugby league matches is associated with team success, greater and longer lasting fatigue and increased injury risk. This study researched the sensitivity and specificity of microtechnology to count collision events during rugby league matches.
Design:
Diagnostic accuracy study.
Methods:
While wearing a microtechnology device (Catapult, S5), eight professional rugby league players were subjected to a total of 380 collision events during matches. Video footage of each match was synchronised with microtechnology data. The occurrence of each collision event was coded in comparison with whether that event was or was not detected by microtechnology.
Results:
Microtechnology detected 371 true-positive collision events (sensitivity=97.6±1.5%). When low-intensity (<1 PlayerLoad AU), short duration (<1s) events were excluded from the analysis, specificity was 91.7±2.5%, accuracy was 92.7±1.3%, positive likelihood ratio was 11.4×/÷1.4 and the typical error of estimate was 7.8%×/÷1.9 (d=0.29×/÷1.9, small). Microtechnology collisions were strongly and positively correlated with video coded collision events (r=0.96). The ability of microtechnology to detect collision events improved as the intensity and duration of the collision increased.
Conclusions:
Microtechnology can identify 97.6% of collision events during rugby league match-play. The typical error associated with measuring contact events can be reduced to 7.8%, with accuracy (92.7%) and specificity (91.7%) improving, when low-intensity (<1 PlayerLoad AU) and short duration (<1s) collision reports are excluded. This provides practitioners with a measurement of contact workload during professional rugby league matches.
The increasing popularity of wearable devices in recent years means that a diverse range of physiological and functional data can now be captured continuously for applications in sports, wellbeing, and healthcare. This wealth of information requires efficient methods of classification and analysis where deep learning is a promising technique for large-scale data analytics. Whilst deep learning has been successful in implementations that utilize high performance computing platforms, its use on low-power wearable devices is limited by resource constraints. In this paper, we propose a deep learning methodology, which combines features learnt from inertial sensor data together with complementary information from a set of shallow features to enable accurate and real-time activity classification. The design of this combined method aims to overcome some of the limitations present in a typical deep learning framework where on-node computation is required. To optimize the proposed method for real-time on-node computation, spectral domain pre-processing is used before the data is passed onto the deep learning framework. The classification accuracy of our proposed deep learning approach is evaluated against state-of-the-art methods using both laboratory and real world activity datasets. Our results show the validity of the approach on different human activity datasets, outperforming other methods, including the two methods used within our combined pipeline. We also demonstrate that the computation times for the proposed method are consistent with the constraints of real-time on-node processing on smartphones and a wearable sensor platform.
