Dataset Information

Recognition of tennis serve performed by a digital player: comparison among polygon, shadow, and stick-figure models.

ABSTRACT: The objective of this study was to assess the cognitive effect of human character models on the observer's ability to extract relevant information from computer graphics animation of tennis serve motions. Three digital human models (polygon, shadow, and stick-figure) were used to display the computationally simulated serve motions, which were perturbed at the racket-arm by modulating the speed (slower or faster) of one of the joint rotations (wrist, elbow, or shoulder). Twenty-one experienced tennis players and 21 novices made discrimination responses about the modulated joint and also specified the perceived swing speeds on a visual analogue scale. The result showed that the discrimination accuracies of the experienced players were both above and below chance level depending on the modulated joint whereas those of the novices mostly remained at chance or guessing levels. As far as the experienced players were concerned, the polygon model decreased the discrimination accuracy as compared with the stick-figure model. This suggests that the complicated pictorial information may have a distracting effect on the recognition of the observed action. On the other hand, the perceived swing speed of the perturbed motion relative to the control was lower for the stick-figure model than for the polygon model regardless of the skill level. This result suggests that the simplified visual information can bias the perception of the motion speed toward slower. It was also shown that the increasing the joint rotation speed increased the perceived swing speed, although the resulting racket velocity had little correlation with this speed sensation. Collectively, observer's recognition of the motion pattern and perception of the motion speed can be affected by the pictorial information of the human model as well as by the perturbation processing applied to the observed motion.

SUBMITTER: Ida H

PROVIDER: S-EPMC3306305 | biostudies-literature | 2012

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Recognition of tennis serve performed by a digital player: comparison among polygon, shadow, and stick-figure models.

Ida Hirofumi H Fukuhara Kazunobu K Ishii Motonobu M

PloS one 20120316 3

The objective of this study was to assess the cognitive effect of human character models on the observer's ability to extract relevant information from computer graphics animation of tennis serve motions. Three digital human models (polygon, shadow, and stick-figure) were used to display the computationally simulated serve motions, which were perturbed at the racket-arm by modulating the speed (slower or faster) of one of the joint rotations (wrist, elbow, or shoulder). Twenty-one experienced te ...[more]

PMID: 22439009

Similar Datasets

Project description:BackgroundThe tennis serve is a complex skill requiring appropriate energy transfer to maximize serve speed. As the only independently powered shot in tennis, it is important to understand what characteristics contribute to a player's serve.HypothesisUpper extremity and lower extremity power variables will be predictive of serve speed.Study designCross-sectional study.Level of evidenceLevel 3.MethodsA total of 42 competitive tennis players underwent range of motion, strength, motor control, power, and serve speed testing. Motion assessment was completed for shoulder external and internal rotation, trunk rotation, hip external and internal rotation, and ankle dorsiflexion. Strength was assessed for the rotator cuff, scapula, and hip musculature. Motor control was evaluated through upper and lower extremity Y-balance testing. Power assessments were made using broad jump, single-leg hop, and seated unilateral shot put testing. Pearson correlation was conducted to assess associations of skill, height, and tested variables to serve speed. Significant variables were used in a stepwise linear regression model, with serve speed as the dependent variable. Variables are listed in relation to the participant's dominant arm.ResultsSkill, height, contralateral hip external rotation range of motion, nondominant arm Y-balance anterolateral reach, bilateral single-leg hops, and seated unilateral shot put throws for both arms demonstrated significant positive correlations to serve speed ( P < 0.05). Serve speed was predicted with 84% variance through skill, height, contralateral hip external rotation range of motion, ipsilateral single-leg hop, and the seated unilateral shot put throws.ConclusionThe ability to generate increased serve speed is multifactorial. The combination of skill, height, hip motion, and upper and lower extremity power may determine serve speed.Clinical relevanceThe findings suggest that motion, motor control, and power testing should be evaluated when working with this population to improve serve speed.

Project description:Human Action Recognition (HAR) is an essential topic in computer vision and artificial intelligence, focused on the automatic identification and categorization of human actions or activities from video sequences or sensor data. The goal of HAR is to teach machines to comprehend and interpret human movements, gestures, and behaviors, allowing for a wide range of applications in areas such as surveillance, healthcare, sports analysis, and human-computer interaction. HAR systems utilize a variety of techniques, including deep learning, motion analysis, and feature extraction, to capture and analyze the spatiotemporal characteristics of human actions. These systems have the capacity to distinguish between various actions, whether they are simple actions like walking and waving or more complex activities such as playing a musical instrument or performing sports maneuvers. HAR continues to be an active area of research and development, with the potential to enhance numerous real-world applications by providing machines with the ability to understand and respond to human actions effectively. In our study, we developed a HAR system to recognize actions in tennis using an attention-based gated recurrent unit (GRU), a prevalent recurrent neural network. The combination of GRU architecture and attention mechanism showed a significant improvement in prediction power compared to two other deep learning models. Our models were trained on the THETIS dataset, one of the standard medium-sized datasets for fine-grained tennis actions. The effectiveness of the proposed model was confirmed by three different types of image encoders: InceptionV3, DenseNet, and EfficientNetB5. The models developed with InceptionV3, DenseNet, and EfficientNetB5 achieved average ROC-AUC values of 0.97, 0.98, and 0.81, respectively. While, the models obtained average PR-AUC values of 0.84, 0.87, and 0.49 for InceptionV3, DenseNet, and EfficientNetB5 features, respectively. The experimental results confirmed the applicability of our proposed method in recognizing action in tennis and may be applied to other HAR problems.

Project description:BackgroundDesigners of digital interventions for mental health often leverage interactions from games because the intrinsic motivation that results from game-based interventions may increase participation and translate into improved treatment efficacy. However, there are outstanding questions about the suitability (eg, are desktop or mobile interventions more appropriate?) and intervention potential (eg, do people with depression activate enough to play?) of games for mental health.ObjectiveIn this paper, we aimed to describe the presently unknown relationship between gaming activity and indicators of well-being so that designers make informed choices when designing game-based interventions for mental health.MethodsWe gathered validated scales of well-being (Beck's Depression Inventory [BDI-II], Patient Health Questionnaire [PHQ-9], trait anxiety [TA], and basic psychological needs satisfaction [BPNS]), play importance (control over game behavior: control; gamer identity: identity), and play behavior (play frequency, platform preferences, and genre preferences) in a Web-based survey (N=491).ResultsThe majority of our participants played games a few times a week (45.3%, 222/490) or daily (34.3%, 168/490). In terms of depression, play frequency was associated with PHQ-9 (P=.003); PHQ-9 scores were higher for those who played daily than for those who played a few times a week or less. Similarly, for BDI-II (P=.01), scores were higher for those who played daily than for those who played once a week or less. Genre preferences were not associated with PHQ-9 (P=.32) or BDI-II (P=.68); however, platform preference (ie, mobile, desktop, or console) was associated with PHQ-9 (P=.04); desktop-only players had higher PHQ-9 scores than those who used all platforms. Platform preference was not associated with BDI-II (P=.18). In terms of anxiety, TA was not associated with frequency (P=.23), platform preference (P=.07), or genre preference (P=.99). In terms of needs satisfaction, BPNS was not associated with frequency (P=.25) or genre preference (P=.53), but it was associated with platform preference (P=.01); desktop-only players had lower needs satisfaction than those who used all platforms. As expected, play frequency was associated with identity (P<.001) and control (P<.001); those who played more had identified more as a gamer and had less control over their gameplay. Genre preference was associated with identity (P<.001) and control (P<.001); those who played most common genres had higher control over their play and identified most as gamers. Platform preference was not associated with control (P=.80), but was with identity (P=.001); those who played on all devices identified more as a gamer than those who played on mobiles or consoles only.ConclusionsOur results suggest that games are a suitable approach for mental health interventions as they are played broadly by people across a range of indicators of mental health. We further unpack the platform preferences and genre preferences of players with varying levels of well-being.

Dataset Information

Recognition of tennis serve performed by a digital player: comparison among polygon, shadow, and stick-figure models.

Publications

Recognition of tennis serve performed by a digital player: comparison among polygon, shadow, and stick-figure models.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets