Optical motion capture dataset of selected techniques in beginner and advanced Kyokushin karate athletes

Marker Data Enhancement For Markerless Motion Capture

Objective

Human pose estimation models can measure movement from videos at a large scale and low cost; however, open-source pose estimation models typically detect only sparse keypoints, which leads to inaccurate joint kinematics. OpenCap, a freely available service for researchers to measure movement from videos, addresses this issue using a deep learning model- the marker enhancer-that transforms sparse keypoints into dense anatomical markers. However, OpenCap performs poorly on movements not included in the training data. Here, we create a much larger and more diverse training dataset and develop a more accurate and generalizable marker enhancer.

Methods

We compiled marker-based motion capture data from 1176 subjects and synthesized 1433 hours of keypoints and anatomical markers to train the marker enhancer. We evaluated its accuracy in computing kinematics using both benchmark movement videos and synthetic data representing unseen, diverse movements.

Results

The marker enhancer improved kinematic accuracy on benchmark movements (mean error: 4.1°, max: 8.7°) compared to using video keypoints (mean: 9.6°, max: 43.1°) and OpenCap's original enhancer (mean: 5.3°, max: 11.5°). It also better generalized to unseen, diverse movements (mean: 4.1°, max: 6.7°) than OpenCap's original enhancer (mean: 40.4°, max: 252.0°).

Conclusion

Our marker enhancer demonstrates both accuracy and generalizability across diverse movements.

Significance

We integrated the marker enhancer into OpenCap, thereby offering its thousands of users more accurate measurements across a broader range of movements.

Kinematic and Physiological Analysis of Medieval Combat Sport Using Motion Analysis, Blood Lactate Measurement, and Heart Rate Monitoring: A Case Study

Medieval combat sport is a form of mixed martial art in which combatants engage in fighting using offensive and defensive equipment while dressed in full armor. The sport is considered extremely taxing, making it nearly impossible to maintain the same level of performance. However, this form of sport has not been thoroughly analyzed, and its impact on human physical response is largely unknown. To address this gap, the study reported here aimed to introduce and test a procedure for analyzing human physical responses within the framework of the sport. To accomplish this, two experienced combatants were asked to engage in a series of strikes, performed in the form of a set duel simulating a professional fight competition. The kinematic aspect of the procedure was examined using motion analysis with the help of an IMU suit, while the physiological aspect was evaluated based on blood lactate levels and heart rate measurements. Furthermore, an ergometer test conducted in a laboratory setting aimed to determine the lactate threshold. The duel results showed noticeable decreases in the kinematic aspects of the strikes, such as the velocity of impact, and a dramatic rise in physiological aspects, such as heart rate and blood lactate levels. During the duel sets, the blood lactate surpassed the threshold level, and at the end, the heart rate exceeded the maximum age-related level. Practicing medieval combat sport has been shown to impose an extreme physical load on the bodies of combatants, noticeably affecting their performance levels.

Development of a Low-Cost Markerless Optical Motion Capture System for Gait Analysis and Anthropometric Parameter Quantification

Technological advancements have expanded the range of methods for capturing human body motion, including solutions involving inertial sensors (IMUs) and optical alternatives. However, the rising complexity and costs associated with commercial solutions have prompted the exploration of more cost-effective alternatives. This paper presents a markerless optical motion capture system using a RealSense depth camera and intelligent computer vision algorithms. It facilitates precise posture assessment, the real-time calculation of joint angles, and acquisition of subject-specific anthropometric data for gait analysis. The proposed system stands out for its simplicity and affordability in comparison to complex commercial solutions. The gathered data are stored in comma-separated value (CSV) files, simplifying subsequent analysis and data mining. Preliminary tests, conducted in controlled laboratory environments and employing a commercial MEMS-IMU system as a reference, revealed a maximum relative error of 7.6% in anthropometric measurements, with a maximum absolute error of 4.67 cm at average height. Stride length measurements showed a maximum relative error of 11.2%. Static joint angle tests had a maximum average error of 10.2%, while dynamic joint angle tests showed a maximum average error of 9.06%. The proposed optical system offers sufficient accuracy for potential application in areas such as rehabilitation, sports analysis, and entertainment.

Accessibility of Motion Capture as a Tool for Sports Performance Enhancement for Beginner and Intermediate Cricket Players

Motion Capture (MoCap) has become an integral tool in fields such as sports, medicine, and the entertainment industry. The cost of deploying high-end equipment and the lack of expertise and knowledge limit the usage of MoCap from its full potential, especially at beginner and intermediate levels of sports coaching. The challenges faced while developing affordable MoCap systems for such levels have been discussed in order to initiate an easily accessible system with minimal resources.

A Systematic Review of Dynamic Forces and Kinematic Indicators of Front and Roundhouse Kicks across Varied Conditions and Participant Experience

Impact force and maximum velocity are important indicators of kick efficiency. Therefore, this systematic review compared the front kick (FK) and roundhouse kick (RK), including their impact force, maximum velocity, angular velocity, and execution time, considering various target types and experience levels. Following PRISMA guidelines, the Web of Science, SportDiscus, and PubMed were systematically searched for articles published from January 1982 to May 2022. Normalized kicking values were compared using one-way ANOVA. Eighteen articles included FKs (sample: 113 elite men, 109 sub-elite men, and 46 novices), and twenty-five articles included RKs (sample: 238 elite men, 143 sub-elite men, and 27 novice men). The results indicate that the impact force of the FK were 47% (p < 0.01), 92% (p < 0.01), and 120% (p < 0.01) higher than those of the RK across novice, sub-elite, and elite groups, respectively. Moreover, the maximum foot velocity of the RK was 44% (p < 0.01) and 48% (p < 0.01) higher than that of the FK for the sub-elite and elite groups, respectively. Furthermore, the elite group had 65% (p < 0.01) higher knee extension angular velocity with the RK than with the FK and 138% (p < 0.01) higher hip extension angular velocity with the FK than with the RK. In summary, the findings suggest that the FK is more effective in generating forceful kicks, while the RK has the potential for rapid execution.

Construction motion data library: an integrated motion dataset for on-site activity recognition

Identifying workers’ activities is crucial for ensuring the safety and productivity of the human workforce on construction sites. Many studies implement vision-based or inertial-based sensors to construct 3D human skeletons for automated postures and activity recognition. Researchers have developed enormous and heterogeneous datasets for generic motion and artificially intelligent models based on these datasets. However, the construction-related motion dataset and labels should be specifically designed, as construction workers are often exposed to awkward postures and intensive physical tasks. This study developed a small construction-related activity dataset with an in-lab experiment and implemented the datasets to manually label a large-scale construction motion data library (CML) for activity recognition. The developed CML dataset contains 225 types of activities and 146,480 samples; among them, 60 types of activities and 61,275 samples are highly related to construction activities. To verify the dataset, five widely applied deep learning algorithms were adopted to examine the dataset, and the usability, quality, and sufficiency were reported. The average accuracy of models without tunning can reach 74.62% to 83.92%.

Personality and Age of Male National Team of Ukraine in Kyokushin Karate-Pilot Study

This article is a continuation of the research on personality in combat sports in karate. The authors’ goal was to verify the relationship between personality and age of kyokushin karate practitioners. The male national team of Ukraine in karate kyokushin (N = 7) participated in the personality study with the use of the Big Five model. The NEO-FFI (NEO Five-Factor Inventory) Personality Questionnaire was applied as a research tool and the package of statistical methods IBM SPSS Statistics 27.0 (IBM Polska, Warszawa, Poland) was used to compute the research results. The study showed that there were differences in the intensity of openness to experiences between individual samples only at the level of the statistical trend. Masters showed a higher level of openness to experiences in relation to juniors (p = 0.081) and seniors (p = 0.097). Also, a negative and strong correlation between the intensity of neuroticism and conscientiousness among the respondents was noted. A conclusion was drawn that, with age, karatekas probably manifest greater openness to experience, which is the result of their sports experience, high sports level and pro-health values of karate. On the other hand, good emotional adaptation of karatekas is strictly related to conscientiousness.

A database of physical therapy exercises with variability of execution collected by wearable sensors

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.

Effects of Core Training in Physical Fitness of Youth Karate Athletes: A Controlled Study Design

This study aimed to analyse the impact of an 8-week core strength training (CST) programme in youth karate practitioners on core endurance, agility, flexibility, sprinting, jumping, and kick performance. This study followed a randomized parallel design. Twenty-nine participants (age: 12.86 ± 0.81 years old; height: 152.55 ± 10.37 cm; weight: 42.93 ± 8.85 kg) were allocated to a CST programme (n = 16) performed thrice weekly or to a control group (n = 13) only performing the sport-specific (karate) training. Participants were assessed three times (baseline, mid and post-intervention) for the following tests: (i) flexor endurance test (FET); (ii) back extensor test (BET); (iii) lateral musculature test (LMT); (iv) flexibility; (v) chance of direction (COD); (vi) countermovement jump (CMJ); (vii) back muscle strength (BMS); (viii) horizontal jump (LJ); (ix) sprint test; and (x) karate kick test (KKT). Between-group analysis revealed significant advantages for the CST group on the FET (p < 0.001), BET (p < 0.001), LMT (p < 0.001), 20 m sprint (p = 0.021) and KKT for right (p < 0.006) and left (p < 0.020) legs. No significant differences were found between groups in the remaining physical fitness variables (p > 0.05). The within-group changes revealed significant improvements in the CST group at flexibility (p = 0.002), COD (p < 0.001), CMJ (p < 0.001), BMS (p < 0.002), 20 m sprint (p = 0.033), and KKT (p < 0.001). In addition, within-group changes in the control group were also significant in flexibility (p = 0.024) and right kick (p < 0.042). We conclude that the CST programme improves core endurance and karate kick performance; however, it is not effective enough for other physical performance parameters in KR practitioners.

Electromyography, Wavelet Analysis and Muscle Co-Activation as Comprehensive Tools of Movement Pattern Assessment for Injury Prevention in Wheelchair Fencing

The aim of the study was to determine the correct movement patterns of fencing techniques in wheelchair fencers. Through a comprehensive analysis, the key muscles in the kinematic chain exposed to potential injuries were identified. The study participants were 16 wheelchair fencers, divided into two groups representing two categories of disability: Group A (N = 7) comprising fencers with mild paraplegia, having freedom of movement of the trunk and arms; and Group B (N = 9) comprising fencers with a spinal cord injury and partial paresis of the arms. EMG and an accelerometer were used as the main research tools. The EMG electrodes were placed on the muscles of the sword arm as well as on the left and right sides of the abdomen and torso. The EMG signal was transformed using wavelet analysis, and the muscle activation time and co-activation index (CI) were determined. In Group A fencers, first the back and abdominal muscles were activated, while in Group B, it was the deltoid muscle. The wavelet coherence analysis revealed intermuscular synchronization at 8–20 Hz for Group A fencers and at 5–15 Hz for Group B fencers. In Group A fencers, the co-activation index was 50.94 for the right-side back and abdominal muscles, 50.75 for the ECR-FCR, and 47.99 for the TRI-BC pairs of upper limb muscles. In contrast, Group B fencers demonstrated higher CI values (50.54) only for the postural left-side muscle pairs. Many overload injuries of the shoulder girdle, elbow, postural muscles, spine, and neck have been found to be preventable through modification of current training programs dominated by specialist exercises. Modern wheelchair fencing training should involve neuromuscular coordination and psychomotor exercises. This will facilitate the individualization of training depending on the fencer’s degree of disability and training experience.

Interjoint Coordination in Kicking a Moving Target: A Comparison Between Elite and Nonelite Taekwondo Players

Patterns of interjoint coordination in the kicking legs of taekwondo players were investigated to understand movement pattern variability as a functional property of skill level. Elite and nonelite players performed roundhouse kicks against a custom-built moving target fitted with an accelerometer, and movements were recorded by motion capture. Average foot segment velocities of 13.6 and 11.4 m/s were recorded for elite and nonelite players, respectively (P < .05), corresponding to target accelerations of 87.5 and 70.8g (P < .05). Gradient values derived from piecewise linear regression of continuous relative phase curves established the comparative incoordination of nonelite taekwondo players in the form of an overshoot behavior during the crucial period leading to target impact (P < .05). This overshoot was apparent in both knee-hip and ankle-knee continuous relative phase curves. Elite players generated greater limb speed and impact force through more effective limb segment coordination. The combination of continuous relative phase and piecewise linear regression techniques allowed identification of alternate joint control approaches in the 2 groups.

Gap Reconstruction in Optical Motion Capture Sequences Using Neural Networks

Optical motion capture is a mature contemporary technique for the acquisition of motion data; alas, it is non-error-free. Due to technical limitations and occlusions of markers, gaps might occur in such recordings. The article reviews various neural network architectures applied to the gap-filling problem in motion capture sequences within the FBM framework providing a representation of body kinematic structure. The results are compared with interpolation and matrix completion methods. We found out that, for longer sequences, simple linear feedforward neural networks can outperform the other, sophisticated architectures, but these outcomes might be affected by the small amount of data availabe for training. We were also able to identify that the acceleration and monotonicity of input sequence are the parameters that have a notable impact on the obtained results.