An approach to sport activities recognition based on an inertial sensor and deep learning

Pająk, Grzegorz; Krutz, Pascal; Patalas‐Maliszewska, Justyna; Rehm, Matthias; Pająk, Iwona; Dix, Martin

doi:10.1016/j.sna.2022.113773

Cited by 23 publications

(10 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The concept of this module is based on the results of our previous research on human activity recognition (HAR). In our previous work [ 7 ], a solution utilizing the technique known as image recognition [ 1 , 3 ] was proposed. The main goal of this work was to find a classifier structure demanding the lowest possible computational effort for the forward pass able to solve the HAR problem.…”

Section: Resultsmentioning

confidence: 99%

“…An important problem that had to be taken into account in the studies presented in this paper was the need to use a classifier for analysis of a live stream of data. In [ 7 , 34 ], the focus was on activity detection, and the input data of the CNN classifier contained only homogeneous signals covering one type of activity in both the training and testing phases. In contrast, in the presented work, the application of a CNN classifier for analysis of ongoing registered signals was considered, so the CNN had to also properly classify heterogenous activity signals.…”

Section: Resultsmentioning

confidence: 99%

“…It is known that applying artificial intelligence (AI) enables a system to support the automation of activities such as identification of correctly/incorrectly performed sport activities [ 5 ]. According to the statement that AI methods can help in training monitoring and training schedule optimization [ 6 ], in previous work [ 7 ], a critical analysis of the literature on this topic was performed from the point of view of convolutional neural network (CNNs) and of CNNs combined with classifiers for HAR applications in sports. In this work, we provide an overview of current solutions that support some feedback features for the correct performance of sports exercises with inertial sensors and CNNs ( Table 1 ).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Inertial Sensor-Based Sport Activity Advisory System Using Machine Learning Algorithms

Patalas‐Maliszewska

Pająk

Krutz

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

The aim of this study was to develop a physical activity advisory system supporting the correct implementation of sport exercises using inertial sensors and machine learning algorithms. Specifically, three mobile sensors (tags), six stationary anchors and a system-controlling server (gateway) were employed for 15 scenarios of the series of subsequent activities, namely squats, pull-ups and dips. The proposed solution consists of two modules: an activity recognition module (ARM) and a repetition-counting module (RCM). The former is responsible for extracting the series of subsequent activities (so-called scenario), and the latter determines the number of repetitions of a given activity in a single series. Data used in this study contained 488 three defined sport activity occurrences. Data processing was conducted to enhance performance, including an overlapping and non-overlapping window, raw and normalized data, a convolutional neural network (CNN) with an additional post-processing block (PPB) and repetition counting. The developed system achieved satisfactory accuracy: CNN + PPB: non-overlapping window and raw data, 0.88; non-overlapping window and normalized data, 0.78; overlapping window and raw data, 0.92; overlapping window and normalized data, 0.87. For repetition counting, the achieved accuracies were 0.93 and 0.97 within an error of ±1 and ±2 repetitions, respectively. The archived results indicate that the proposed system could be a helpful tool to support the correct implementation of sport exercises and could be successfully implemented in further work in the form of web application detecting the user’s sport activity.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Inertial Sensor-Based Sport Activity Advisory System Using Machine Learning Algorithms

Patalas‐Maliszewska

Pająk

Krutz

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…In addition to the above classification algorithms (SVM [72][73][74][75]43,61], GMM [76,77], HMM [70,69,78]), some other classification algorithms are used for posture recognition, such as k-nearest neighbor (k-NN) [79], random forest (RF) [80][81][82], Bayesian classification algorithm [83], decision tree (DT) [72,84,85], linear discriminant analysis [86,60], naïve Bayes (NB) [72,87], etc.…”

Section: Other Classification Approachesmentioning

confidence: 99%

A Survey on Artificial Intelligence in Posture Recognition

Jiang¹,

Hu²,

Wang³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Over the years, the continuous development of new technology has promoted research in the field of posture recognition and also made the application field of posture recognition have been greatly expanded. The purpose of this paper is to introduce the latest methods of posture recognition and review the various techniques and algorithms of posture recognition in recent years, such as scale-invariant feature transform, histogram of oriented gradients, support vector machine (SVM), Gaussian mixture model, dynamic time warping, hidden Markov model (HMM), lightweight network, convolutional neural network (CNN). We also investigate improved methods of CNN, such as stacked hourglass networks, multi-stage pose estimation networks, convolutional pose machines, and high-resolution nets. The general process and datasets of posture recognition are analyzed and summarized, and several improved CNN methods and three main recognition techniques are compared. In addition, the applications of advanced neural networks in posture recognition, such as transfer learning, ensemble learning, graph neural networks, and explainable deep neural networks, are introduced. It was found that CNN has achieved great success in posture recognition and is favored by researchers. Still, a more in-depth research is needed in feature extraction, information fusion, and other aspects. Among classification methods, HMM and SVM are the most widely used, and lightweight network gradually attracts the attention of researchers. In addition, due to the lack of 3D benchmark data sets, data generation is a critical research direction.

show abstract

“…Due to the need of fairness and justice in the examination, the system is required to have accurate judgment, and the following requirements should be realized in the examination: automatically capture the first drop of the volleyball 1 ; Automatically determine whether the first landing point is in the scoring zone; Scores are divided into: in-bounds and out-of-bounds (9 m ×9 m half-court area);The test subjects continuously spike the ball and serve, and the decision result is given 3 seconds after each ball falls; Landing site identification accuracy ≥99.9%;The accuracy rate of area recognition of landing site is ≥99.9%;Landing position error;±10mm;Ground frame recognition error;±1 frame; Extended function: can customize the partition.…”

Section: Technical Requirements Of National Entrance Examinationmentioning

confidence: 99%

Artificial intelligence volleyball special test system based on deep learning and 3D imaging technology

Li,

Xiong

2023

Fifth International Conference on Artificial Intelligence and Computer Science (AICS 2023)

View full text Add to dashboard Cite

With the rise of digitalization and intelligence, more and more electronic products begin to replace manual labor, "artificial intelligence volleyball special examination system" is a combination of computer deep learning and three-dimensional imaging technology of college entrance examination service system. The development and application of this system is intended to replace the traditional manual scoring mode, open up a new path for the modern examination mode, and further ensure the fairness of the examination. In this paper, sports literature, experiment, data-driven and other research methods, constantly learn and record volleyball drop point, ball speed and other data in the actual field such as exams and competitions, analyze the data after data collection, and learn and discover patterns, rules and knowledge from the data through training models. Research and development was carried out through data results, and through in-depth calculation, the computer finally reached the point where it could identify various signs in volleyball, volleyball court and volleyball court. The hawk-eye camera could capture the trajectory of volleyball movement in real time without dead corners, and the computer controlled deep learning of volleyball movement images to automatically score the result according to different landing areas. The system through continuous learning, analysis, calculation and other research and development links to achieve accurate prediction effect, and its application in the exam, has achieved accurate points instead of manual, fair and fast, the application of manual system in modern exams has been very necessary, has become an indispensable part of the modern exam.

show abstract

An approach to sport activities recognition based on an inertial sensor and deep learning

Cited by 23 publications

References 18 publications

Inertial Sensor-Based Sport Activity Advisory System Using Machine Learning Algorithms

Inertial Sensor-Based Sport Activity Advisory System Using Machine Learning Algorithms

A Survey on Artificial Intelligence in Posture Recognition

Artificial intelligence volleyball special test system based on deep learning and 3D imaging technology

Contact Info

Product

Resources

About