Using complex networks and multiple artificial intelligence algorithms for table tennis match action recognition and technical-tactical analysis

The main core purpose of artificial emotional intelligence is to recognize human emotions. Technologies such as facial, semantic, or brainwave recognition applications have been widely proposed. However, the abovementioned recognition techniques for emotional features require a large number of training samples to obtain high accuracy. Human behaviour pattern can be trained and recognized by the continuous movement of the Spatial Temporal Graph Convolution Network (ST-GCN). However, this technology does not distinguish between the speed of delicate emotions, and the speed of human behaviour and delicate changes of emotions cannot be effectively distinguished. This research paper proposes Spatial Temporal Variation Convolutional Network training for human emotion recognition, using skeleton detection technology to calculate the degree of skeleton point change between consecutive actions and using the nearest neighbour algorithm to classify speed levels and train the ST-GCN recognition model to obtain the emotional state. Application of the speed change recognition ability of the Spatial Temporal Variation Graph Convolution Network (STV-GCN) to artificial emotional intelligence calculation makes it possible to efficiently recognize the delicate actions of happy, sad, fear, and angry in human behaviour. The STV-GCN technology proposed in this paper is compared with ST-GCN and can effectively improve the recognition accuracy by more than 50%.

Spatial cognition facilitates the successful completion of specific cognitive tasks through lateral processing and neuroplasticity. Long-term training in table tennis induces neural processing efficiency in the visuospatial cognitive processing cortex of athletes. However, the lateralization characteristics and neural mechanisms of visual–spatial cognitive processing in table tennis players in non-sport domains are unclear. This study utilized event-related potentials to investigate differences in the spatial cognition abilities of regular college students (controls) and table tennis players. A total of 48 participants (28 controls; 20 s-level national table tennis players) completed spatial cognitive tasks while electroencephalography data were recorded. Task performance was better in the table tennis group than in the control group (reaction time: P < 0.001; correct number/sec: P = 0.043), P3 amplitude was greater in the table tennis group (P = 0.040), spatial cognition showed obvious lateralization characteristics (P < 0.001), table tennis players showed a more obvious right-hemisphere advantage, and the P3 amplitude in the right hemisphere was significantly greater in table tennis athletes than in the control group. (P = 0.044). Our findings demonstrate a right-hemisphere advantage in spatial cognition. Long-term training strengthened the visual–spatial processing ability of table tennis players, and this advantage effect was reflected in the neuroplasticity of the right hemisphere (the dominant hemisphere for spatial processing).

An interesting and little-reported problem in the literature is the scale of asymmetry in table tennis players, the magnitude of which should perhaps be treated as a risk for injury. Determining the degree of asymmetry in table tennis players can indicate the need to appropriately manage the training process, including compensatory or corrective exercises in the training program, especially since recent studies confirm that training interventions can reduce sporting asymmetries and improve performance. This study aimed to assess the amount of asymmetry in the trunk regarding the frontal plane and the difference between limb circumferences in female table tennis players compared to the control group (non-athletes). Twenty-two women took part in the study. Ten of them were table tennis professionals with an average training experience of 7±4.3 years (the exclusion criterion of the study was a minimum of 3 years of training experience). As a comparison group, the study included 12 female students who did not participate in competitive sports. Body posture was assessed in all subjects using equipment for computer analysis of asymmetry in the torso using the photogrammetric method. Additionally, all the subjects had their upper and lower limb circumferences measured. The results of the conducted research showed asymmetry in the frontal plane in the table tennis player group. As many as six parameters-regarding the pelvic rotation angle, angle of trunk inclination, the height of the angles of the lower shoulder blades and their distance from the spine, as well as the waist triangles, difference in the width and height of the waist triangles and the angle of trunk inclination-indicated asymmetry in this group but significantly differed from the control group (p≤0.05) only in the first parameter given above. The calculated differences in circumference between the right and left sides in the individual groups were statistically different in several cases (p≤0.05). This concerned the circumferences of the arms, forearms, elbows, and knees of table tennis players. The research carried out in this study allowed us to determine the occurrence of asymmetry in the frontal plane of the trunk and between the limbs of table tennis players. According to some studies, this may be a risk factor for injury. However, despite the lack of uniform views in the literature on the importance and threats resulting from asymmetries, it appears that, if only for aesthetic reasons, table tennis would require compensatory or corrective training aimed at developing symmetry of the body structure.

Table tennis players execute short explosive movements, along with continuous hip flexion, abduction, and rotation, increasing their risk of injury. Previous studies reported a rate of return to sports (RTS) of 20% to 80% in athletes following total hip arthroplasty (THA). There are no studies reporting RTS in table tennis players following THA. To evaluate the clinical outcomes and RTS following custom THA in professional, ex-professional, and recreational table tennis players. Case series; Level of evidence, 4. Patients who underwent primary THA between April 2013 and January 2022 were retrospectively reviewed (n = 2977). Table tennis players of any level that received a custom femoral stem were included in the study (N = 17). At a minimum follow-up of 2 years, all players were assessed using the Oxford Hip Score (OHS), Forgotten Joint Score (FJS), and the University of California Los Angeles (UCLA) activity score, as well as with a sports-specific questionnaire that included questions related to their table tennis practice. Descriptive statistics, including medians and interquartile ranges, were used to summarize the data. All 17 players (22 hips) were available at ≥2 years, of which 3 were professional (5 hips), 4 were ex-professional (6 hips), and 10 were recreational (11 hips). The median OHS was 44.0 (IQR, 44.0-48.0) in professional, 48.0 (IQR, 48.0-48.0) in ex-professional, and 48.0 (IQR, 45.0-48.0) in recreational players. The median FJS was 92.0 (IQR, 88.0-98.0) in professional, 98.0 (IQR, 98.0-98.0) in ex-professional, and 100.0 (IQR, 93.0-100.0) in recreational players. The median UCLA activity score was 10.0 (IQR, 9.0-10.0) in professional, 9.0 (IQR, 9.0-9.8) in ex-professional, and 8.0 (IQR, 5.5-9.0) in recreational players. The rate of RTS was 100% for professional and ex-professional players, and 80% for recreational players. The hours played before onset of symptoms was higher than following surgery for professional (30.0 [IQR, 25.0-30.0] vs 20.0 [IQR, 16.0-22.5] h/week) and ex-professional players (19.5 [IQR, 11.0-29.3] vs 3.0 [IQR, 2.0-5.5] h/week), while it was constant for recreational players (4.0 [IQR, 2.3-4.0] vs 4.0 [IQR, 3.8-4.5] h/week). Our retrospective analysis demonstrated that at a minimum follow-up of 2 years THA using custom stems provided good to excellent clinical outcomes in professional, ex-professional, and recreational table tennis players. All professional and ex-professional players, as well as 80% of recreational players, were able to return to play table tennis, although both professional and ex-professional players reduced their number of hours of play compared with before surgery. These findings could be used to help set expectations for table tennis players who are scheduled to undergo THA.

ABSTRACTTalent detection intends to support lifelong sports participation, reduce dropouts and stimulate sports at the elite level. For this purpose it is important to reveal the specific profile which directs children to the sports that connect to their strengths and preferences. This study evaluated a perceptuomotor skills assessment as part of talent detection for table tennis, a sport in which perceptuomotor skills are considered essential to cope with the difficult technical aspects. Primary school children (n = 121) and gifted young table tennis players (n = 146) were assessed using the Dutch perceptuomotor skills assessment measuring “ball control” and “gross motor function”. A discriminant function analysis confirmed the added value by identifying primary school children fitting the table tennis perceptuomotor profile of the young gifted table tennis players (28%). General linear model analyses for the assessment’s individual test items showed that the table tennis players outperformed their primary school peers on all “ball control” items (P < 0.001). In conclusion, the assessment appears to be of added value for talent detection in table tennis at this young age. Longitudinal studies need to reveal the predictive value for sports participation and elite sports.

The attribute of speed serves the motor performance and gives it the required effectiveness, and the attributes of strength and speed together serve the motor performance so that it performs properly and effectively, thus the ability to score points in the shortest possible time and win against the competitor. The importance of the research lies in preparing special exercises to develop attention aspects and the forehand and backhand skills in table tennis for players according to the correct performance of the skill. From this standpoint, the researcher turned to this study, believing that these special exercises will contribute to developing the physical and skill levels of table tennis players. The research problem was noticed by the researchers through their previous experiences as practitioners of the game in the weakness of attention aspects among table tennis players. The researcher believes that the reason for this is that most coaches do not use modern devices during their training for skills, as they usually rely on traditional methods in implementing and teaching the skill to beginners. Therefore, the researcher decided to add some modern techniques during her study of this problem by trying to use special exercises to develop attention aspects and some table tennis skills for players. The research aims to prepare special exercises to develop attention aspects and some table tennis skills for players, and to identify the effect of special exercises in developing attention aspects and some table tennis skills for players, Conclusions, the results showed that there is a positive effect of special exercises in developing attention aspects (focus – transfer – distribution) among table tennis players aged (12-14) years. The results also showed that there is a positive effect of special exercises in developing some table tennis skills for players aged (12-14) years.

Action recognition, a field within human-centered computing, is vital for identifying and understanding human actions, benefiting applications like surveillance, autonomous vehicles, and human-computer interaction. Artificial intelligence’s enduring goal is to develop robust models for perceiving and understanding the visual world around us. Deep neural networks have shown exceptional performance in various tasks, profoundly impacting real-world applications, including action prediction and recognition, which have advanced significantly in recent years. Most work in visual human action recognition focuses on single viewpoints or modalities from complete observation. Yet, the real significance lies in predicting future actions from incomplete observations to prevent real-world tragedies. With the availability of multiple cameras and data from multiple modalities (RGB, Depth, and Skeleton) available today, it becomes possible to model human action in multi-view and multi-modality context, minimizing the data loss due to occlusions and signal quality issues to improve recognition accuracy on the strength of state-of-the-art deep learning models. However, deep neural network models are susceptible to adversarial attacks, where imperceptible perturbations can compromise action recognition model performance. This thesis focuses on identifying latent vulnerabilities and proposing a defense mechanism against such threats in a multi-modality and multi-view setting. This work introduces an efficient and effective attack mechanism that perturbs skeleton data by targeting key joints and segments while employing a graph attention mechanism that learns the semantics to perturb other modalities. Additionally, an approach has been developed that not only adds noise but also alters the visual spatial structure of skeleton data through generative modeling. Furthermore, this dissertation introduces a defense mechanism known as the Collaborative Knowledge Distillation Network, which leverages graph attention and knowledge distillation techniques. This network leverages the knowledge from compromised multi-view data and integrates information from clean data to address incomplete observations and noisy action videos, enhancing the robustness of action recognition models for real-world applications.

In this paper, we address the problem of cross-view action recognition from a monocular RGB camera. This topic has been considered extremely challenging due to the lack of 3D information in 2D images. Exploiting the advances in 3D pose estimation from a single RGB camera, we propose a new framework termed DeepVI, for cross-view action recognition without the need for pose alignment. Virtual viewpoints are used to augment the variability of training data along with the use of an end-to-end Deep Neural Network (DNN). The proposed network is composed of two modules. The first one, called SmoothNet, implicitly smooths skeleton joint trajectories using revisited temporal convolution in order to reduce the noise in the estimated 3D skeletons. The second module consists of a state-of-the-art approach designed for action recognition based on Spatial Temporal Graph Convolutional Networks (ST-GCN [40]). Experiments have been conducted in cross-view settings on two datasets, namely, NTU RGB-D and Northwestern-UCLA. The obtained results show the effectiveness of the proposed framework.

Dynamics of human body skeletons convey significant information for human action recognition. Conventional approaches for modeling skeletons usually rely on hand-crafted parts or traversal rules, thus resulting in limited expressive power and difficulties of generalization. In this work, we propose a novel model of dynamic skeletons called Spatial-Temporal Graph Convolutional Networks (ST-GCN), which moves beyond the limitations of previous methods by automatically learning both the spatial and temporal patterns from data. This formulation not only leads to greater expressive power but also stronger generalization capability. On two large datasets, Kinetics and NTU-RGBD, it achieves substantial improvements over mainstream methods.

The purpose of this study was to examine the field perception and outcome of competition in order to promote training and competition effects in table tennis players. Using convenient sampling, this study sampled data from 50 players of men's single table tennis from the 2015 Taiwan Collegiate Games. This study recorded every match and asked players to complete the field perception after competition. With χ^2 and logistic regression analysis, this study found the following results: (1) there was a significant relationship between field perception and outcome of competition in pre-serve; (2) during the tie situation, field perception was correlated with competition outcome; (3) the field perception after outcome of a score was correlated with next score; and (4) field perception shown to promote sport performance in table tennis players. This study concludes that the field perception in pre-serve, tie situation and outcome of last score were significantly correlated with the next score. These results indicate that field perception is beneficial to competition outcome for the table tennis players. This study suggests table tennis players and coaches should strengthen positive field perception in competition to produce more positive perception under critical conditions, which certainly help to improve the performance and win the competitions.

ABSTRACTTable tennis is a sport in which players perform technical actions at high speed. In this regard, response time (TTR), reaction time (RT) and movement time (MT) are key factors that can contribute to successful performance of table tennis players. The purposes of this study were to determine the magnitude of intra-subject and inter-subject variability in TTR based on player’s lateral dominance, to establish possible differences between sex, age group and ranking position for TTR, RT and MT, and to relate these results with table tennis performance by analysing the achieved position in the national ranking. A total of 116 Spanish national level players (71 males and 45 females) carried out the Take-Off Reaction Test in a table tennis court simulating real sport conditions. There were no intra-sex differences for the total sample when lateral dominance was checked. On the other hand, male players showed lower MT than females whereas female players showed significantly lower RT values than males. Moreover, and taken into account the correlation analysis, we can conclude that both lower RT and MT can be considered as key variables of performance in table tennis.

The recognition of sports action is an important research subject, which is conducive to the improvement of athletes’ own level. To improve the accuracy of multi-modal data action recognition, based on the Transformer module, this study introduces a multi-head attention mechanism, fuses multi-modal data, and constructs a multi-stream structured object relationship inference network. Based on PointNet++ network and combining five different data fusion frameworks, a motion recognition model that integrates RGB data and 3D skeleton point cloud is constructed. The results showed that the Top-1 accuracy of multi-stream structured object relationship inference network was 42.5% and 42.7%, respectively, which was better than other algorithms. The accuracy of the multi-modal fusion model was improved by 15.6% and 5.1% compared with the single mode, and by 5.4% and 2.6% compared with the dual mode, which showed its superiority in the action recognition task. This showed that the fusion of multi-modal data can provide more abundant information, so as to improve the accuracy of action recognition. The accuracy of the action recognition model combining RGB data and 3D skeleton point cloud was 84.3%, 87.5%, 90.2%, 90.6% and 91.2% after the combination of different strategies, which effectively compensated for the problem of missing information in 3D skeleton point cloud and significantly improved the accuracy of action recognition. With a small amount of data, the Top-1 accuracy of the multi-stream structured object relationship inference network in this study was superior to other algorithms, showing its advantages in dealing with complex action recognition tasks. In addition, the action recognition model that fuses RGB data and 3D skeleton point cloud also achieved higher accuracy, which is better than other algorithms. This study can meet the needs of motion recognition in different scenarios and has certain reference value.

The impact of negative list policy on sectoral structure: Based on complex network and DID analysis

Using Artificial intelligence to assist sports training is a hallmark of the information age. Table tennis, a popular sport in China and worldwide, can also benefit from the latest technological advancements in learning and training. Objective: This study aims to quantitatively evaluate the effectiveness of forehand strokes of different-level table tennis players using dynamic time-warping (DTW) algorithms in AI. Twenty-five beginner players (no prior Table Tennis experience, Male:22 Female:3, Age:18-28) and twenty-five intermediate players (At least 2 years of Table Tennis experience, Male:22 Female:3, Age:18-28) were recruited based on convenience sampling. After obtaining consent, they received 2 hours of basic training on the forehand stroke. Then, their hitting poses are recorded with the system for a 15-second video stream. A dynamic time-wrapping (DTW) algorithm was applied to recognize individual strokes and calculate the difference between the elbow angles of beginner players’ strokes and intermediate players’ strokes; the DTW value was the square of the difference between the two angles. At the same time, the recording is sent to two professional table tennis coaches to evaluate the hitting position with a composite score of 0-100, where 100 stands for most effective and 0 stands for least effective. And then compare their hitting poster with professional athletes. The evaluation criteria included good timing, speed, landing point, and movement rationalization. A study of 25 beginners and 25 intermediate table tennis players analyzed stroke effectiveness via two coach evaluations (0-100) and normalized DTW values. Coaches demonstrated high inter-rater reliability (beginners: p=0.48; intermediates: p=0.72) and significantly distinguished skill levels (p &lt; 0.001). Beginners exhibited higher DTW values (48.14±10.81 vs. 40.17±7.34, p=0.004) with a large Cohen’s d 0.86, indicating a substantial difference. These findings confirm DTW’s capacity to statistically differentiate skill levels, aligning with expert assessments. Thus, normalized DTW effectively correlates with human-evaluated stroke quality, validating its utility as an objective metric for distinguishing player proficiency. Discussion/Conclusion: With a dynamic time-wrapping algorithm, the system could differentiate beginner and intermediate players. One area for improvement lies in refining the DTW algorithm to account for a broader range of individual factors such as body height, arm span, and player-specific biomechanics. Also, the DTW score can potentially be used to rate performance in terms of a player’s posture, grip, swing mechanics, and other racket sports.

Total daily energy expenditure (TEE) and physical activity level (PAL) are important for adequate nutritional management in athletes. The PAL of table tennis has been estimated to about 2.0: it is categorized as a moderateactivity sport (4.0 metabolic equivalents [METs]) in the Compendium of Physical Activities. However, modern table tennis makes high physiological demands. The aims of the current study were to examine (1) TEE and PAL of competitive table tennis players and (2) the physiological demands of various types of table tennis practice. In Experiment 1, we measured TEE and PAL in 10 Japanese college competitive table tennis players (aged 19.9 ± 1.1 years) using the doubly labeled water (DLW) method during training and with an exercise training log and self-reported energy intake. TEE was 15.5 ± 1.9 MJ·day-1 (3695 ± 449 kcal·day-1); PAL was 2.53 ± 0.25; and the average training duration was 181 ± 38 min·day-1. In Experiment 2, we measured METs of five different practices in seven college competition players (20.6 ± 1.2 years). Three practices without footwork were 4.5-5.2 METs, and two practices with footwork were 9.5-11.5 METs. Table tennis practices averaged 7.1 ± 3.2 METS demonstrating similarities with other vigorous racket sports. In conclusion the current Compendium of Physical Activities underestimates the physiological demands of table tennis practice for competition; the estimated energy requirement should be based on DLW method data.