Search

Human activity recognition is known as the backbone of the development of interactive systems, such as computer games. This process is usually performed by either vision-based or depth sensors. So far, various solutions have been developed for this purpose; however, all the challenges of this process have not been completely resolved. In this paper, a solution based on pattern recognition has been developed for labeling and scoring physical exercises performed in front of the Kinect sensor. Extracting the features from human skeletal joints and then generating relative descriptors among them is the first step of our method. This has led to quantification of the meaningful relationships between different parts of the skeletal joints during exercise performance. In this method, the discriminating descriptors of each exercise motion are used to identify the adaptive kernels of the Constrained Energy Minimization method as a target detector operator. The results indicated an accuracy of 95.9% in the labeling process of physical exercise motions. Scoring the exercise motions was the second step after the labeling process, in which a geometric method was used to interpolate numerical quantities extracted from descriptor vectors to transform into semantic scores. The results demonstrated the scoring process coincided with the scores derived by the sports coach by a 99.5 grade in the R2 index. [ABSTRACT FROM AUTHOR]

Language is a key resource for speakers and signers to index different aspects of their social identities, such as their ethnicities and sexualities. Yet, for users of sign language – who exploit movements of the hands, face, head and torso for linguistic purposes – it is often assumed that any communicative movement of the body is part of a sign language rather than a general feature of the body's potential to communicate social meaning, shared by hearing and deaf individuals. In this study, we test this claim by comparing the movement features produced by gay and straight Israeli Sign Language signers to the gestural movements produced by gay and straight Hebrew speakers. The findings reveal that deaf gay signers and hearing gay gesturers exploit similar movements of the body. By incorporating the notion of embodiment into sign language sociolinguistics, we can better conceptualize the relationship between sign language and social identities. [ABSTRACT FROM AUTHOR]

Introduction: Our recent reports have shown that movement pattern and displacement assessed with an infrared depth sensor during a 20-s stepping-in-place test (ST) correlates with measures of balance and need for assistance in older individuals. This study investigated ST test-retest reliability in community-dwelling older adults with and without supportive care. Methods: Two groups were included: those not participating (HO: n = 25, 74.7 ± 5.2 years) and those participating (DSU: n = 41, 78.8 ± 5.8 years) in regular senior day services (supportive care). HO completed three ST trials, separated by 1 week, while DSU completed two ST trials during the same half-day supportive care visit. Testing was conducted with eyes open. ST measures included head movement path distance (TMD), maximum movement displacement (MMD), knee movement path length (KMD), and stepping rate (STEP). TMD×KMD−1 ratio indicated upper-body sway relative to lower-body. Test-retest reliability (intra-class correlation coefficients, ICCs) and between-trial and between group differences (ANCOVA, adjusting for age) were assessed. Results: After adjusting for age, TMD, KMD, TMD×KMD−1 were lesser and STEP was greater in HO than DUS. HO ST variables did not differ across testing days. HO ICC (1, 3) for TMD (0.911 [95% confidence interval: 0.827–0.958]), MMD (0.918 [0.814–0.961]), KMD (0.838 [0.685–0.923]), and TMD×KMD−1 (0.940 [0.884–0.872]) showed strong to very strong reliability. Similarly, DSU ST variables did not differ across same day trials and ICC (1, 2) for TMD, KMD, and TMD×KMD−1 displayed very strong reliability. Conclusion: These ST variables exhibited excellent test-retest reliability of discriminating between community-dwelling older adults with and without supportive care. [ABSTRACT FROM AUTHOR]

Everyone aspires to live a healthy life, but many will inevitably experience some form of disease, illness, or accident that results in disability at some point. Rehabilitation plays a crucial role in helping individuals recover from these disabilities and return to their daily activities. Traditional rehabilitation methods are often expensive, are inefficient, and lead to slow progress for patients. However, in this era of technology, various sensor-based automatic rehabilitation is also possible. A Kinect sensor is a skeletal tracking device that captures human motions and gestures. It can provide feedback to the users, allowing them to better understand their progress and adjust their movements accordingly. In this study, stroke-based rehabilitation is presented along with the Toronto Rehab Stroke Pose Dataset (TRSP). Pre-processing of the raw dataset was performed using various features, and several state-of-the-art classifiers were applied to evaluate the data provided by the Kinect sensor. Among the various classifiers, eXtreme Gradient Boosing (XGB) attained the maximum accuracy of 92% for the TRSP dataset. Furthermore, hyperparameters of the XGB have been optimized using a metaheuristic gray wolf optimizer for better performance. [ABSTRACT FROM AUTHOR]

Industry 5.0 emphasises collaborative work between humans, advanced technology, and artificial intelligence robots, focusing on human-centric principles and integrating flexibility and sustainability to enhance workflows. The advancement of human-robot interaction services can significantly improve the operations efficiency of digital twin manufacturing cells. Motivated by the above background, a gesture-driven interaction architecture for digital twin manufacturing cells is proposed, including data acquisition layer, data processing layer, and application service layer. Secondly, deep learning algorithms are employed for recognising predefined gestures. An improved YOLOv5 algorithm is used to solve the problem of low accuracy in static gesture recognition; while a 3D-CNN-based multimodal data fusion algorithm is used to solve the problem of continuity, diversity, and dimensionality in dynamic gesture recognition. Ultimately, the prototype system is developed utilising Kinect 2.0 and Unity 3D, which involves linking the gesture recognition to the digital twin model, and linking the digital twin model to the physical manufacturing cells. This study is expected to provide theoretical and practical insights to empower human-robot interaction technology in manufacturing cells. [ABSTRACT FROM AUTHOR]

In order to enable rescue robots to accurately detect human joint points in dangerous building environments to replace rescuers into dangerous scenes. A machine learning based human joint point detection method was proposed. Firstly, human joint detection data was obtained through Kinect cameras, and normalized methods were used to smooth and preprocess the joint data. Secondly, human joint length constraints and joint rotation angle constraints were added to the preprocessed data to generate the spatial positions of each human joint point. Finally, the Tsai method was used for hand eye calibration to obtain the distance between the robotic arm and the injured person. The experimental results show that the method accurately measures the position of the casualty joint point in the rescue environment of a dangerous building, which verifies the effectiveness of the proposed detection method. [ABSTRACT FROM AUTHOR]

This study aimed to identify quantitative biomarkers of motor function for cerebellar ataxia by evaluating gait and postural control using an RGB-depth camera-based motion analysis system. In 28 patients with degenerative cerebellar ataxia and 33 age- and sex-matched healthy controls, motor tasks (short-distance walk, closed feet stance, and stepping in place) were selected from a previously reported protocol, and scanned using Kinect V2 and customized software. The Clinical Assessment Scale for the Assessment and Rating of Ataxia (SARA) was also evaluated. Compared with the normal control group, the cerebellar ataxia group had slower gait speed and shorter step lengths, increased step width, and mediolateral trunk sway in the walk test (all P < 0.001). Lateral sway increased in the stance test in the ataxia group (P < 0.001). When stepping in place, the ataxia group showed higher arrhythmicity of stepping and increased stance time (P < 0.001). In the correlation analyses, the ataxia group showed a positive correlation between the total SARA score and arrhythmicity of stepping in place (r = 0.587, P = 0.001). SARA total score (r = 0.561, P = 0.002) and gait subscore (ρ = 0.556, P = 0.002) correlated with mediolateral truncal sway during walking. These results suggest that the RGB-depth camera-based motion analyses on mediolateral truncal sway during walking and arrhythmicity of stepping in place are useful digital motor biomarkers for the assessment of cerebellar ataxia, and could be utilized in future clinical trials. [ABSTRACT FROM AUTHOR]

Traditional robotic arms rely on complex programming and predefined trajectories to operate, which limits their applicability. To improve the flexibility and adaptability of the robot arm, the research focuses on improving the grasping performance of the robot arm based on vision technology. Kinect technology is used to capture human arm movements, and Kalman filter is introduced to smooth image data, so as to optimize the motion recognition process. In this study, the residual network model is further improved, and ELU activation function and pre-activation mechanism are introduced to enhance the classification accuracy of gesture images. The results showed that the improved ResNet50 model achieves 95% recognition accuracy after 25 iterations of training, while the original model is 80%. The application of Kalman filter makes the motion tracking curve smoother and shows the correction effect of this method. In simulation tests, the robotic arm is able to identify different elbow bending angles with 90-96 percent accuracy, while mimicking five specific hand gestures with 96-98 percent accuracy. These data support the practicability and effectiveness of the application of vision capture technology and deep learning model in the field of intelligent control of robotic arms. [ABSTRACT FROM AUTHOR]

Traditional method of teaching poses two significant problems - not all students learn alike, and the physical interaction needed poses health risk during pandemic. As such, for these students, an alternative learning method such as those that uses natural user interface (NUI) can be considered. This method would be beneficial for kinesthetic type learners and can be conducted remotely. The alternative learning program is a complementary method, thus still incorporates the current subject syllabus. However, the delivery, learning and execution of the syllabus will be varied. In minimizing these gaps found in the current Malaysian education system, a conceptual framework utilizing Microsoft Kinect is proposed. Since this is a new framework, a pilot study is needed to gauge the validity and reliability of the survey instrument prior to embarking on further study on the outcome of the alternative learning program. Face and content validity conducted on the questionnaire were found to be clear, not confusing, and measures what the questions were supposed to measure. Reliability measured using Cronbach's Alpha indicated values above the acceptable range. Thus, these results indicate that the instrument is valid and reliable to be applied for data collection in the future study to assess the intention of Malaysian students to adopt an alternative medium for learning. [ABSTRACT FROM AUTHOR]

As manufacturing processes of complex products become automated, manual operations still occupy a considerable portion of industrial maintenance and assembly (IMA), especially in the machinery and aerospace fields. Workers often exhibit awkward posture in IMA activities. In these scenarios, posture assessment is critical for improving the well-being of workers because awkward postures can lead to work-related musculoskeletal disorder (WMSD). Although there are several categorized WMSD risk assessment methods, limited evidence suggests that these methods are compliant for modern complex IMA scenarios. In this paper, a posture analysis system for manual operation (PASMO) is presented to monitor working postures and evaluate WMSD risks in IMA processes. The noninvasive depth sensor Kinect v2 and the rapid upper limb analysis (RULA) method are integrated to achieve this purpose. In the PASMO, the RULA is optimized and driven by motion capture (MoCap) data to make evaluating the WMSD risk of working postures more effective and accurate. Industrial and laboratory experiments are designed to verify the effectiveness and system performance of the PASMO. The results show that for the overall body and most joints, the scores obtained by the PASMO substantially agree with those obtained by the ground truth data (p < 0.01, κ = 0.65) under the real industrial environment. Because the experiments are conducted in real IMA scenarios with body occlusion, the results preliminarily prove the effectiveness of the PASMO for WMSD risk assessment in continuous IMA tasks. [ABSTRACT FROM AUTHOR]

In today's digital era, human-computer interaction interfaces evolve and increase together with the needs of the users. However, the existing technologies have their limitations, which can hinder the efficiency of modern input devices like the Kinect sensor or other similar sensors. In this paper we improved our previous algorithm by extending it with two algorithms that aim to help telerehabilitation for individuals with movement disabilities. These two algorithms are based on the k-Nearest Neighbors, the k-Means algorithms. The algorithms are designed to accommodate the needs of the patients by adapting to their gestures based on their previous three. Using these gestures, the algorithms create multiple gesture acceptance domains around each coordinate of the gesture. Consequently, they decide whether the next user-input gesture can be considered the same movement. The accuracy of these algorithms was evaluated in three acceptance domains by comparing gesture descriptors with either the Euclidean or the Manhattan distance calculation methods. The results show that k-Nearest Neighbors algorithm yields better results in larger acceptance domains, while the k-Means algorithm can provide a better gesture acceptance rate in the smaller ones. The results show that both algorithms can be used in the telerehabilitation process, although the k-Means algorithm is more accurate than the k-Nearest Neighbors algorithm. [ABSTRACT FROM AUTHOR]

Recently, markerless tracking systems, such as RGB-Depth cameras, have spread to overcome some of the limitations of the gold standard marker-based tracking systems. Although these systems are valuable substitutes for human motion analysis, as they guarantee higher flexibility, faster setup time and lower costs, their tracking accuracy is lower with respect to marker-based systems. Many studies quantified the error made by markerless systems in terms of body segment length estimation, articular angles, and biomechanics, concluding that they are appropriate for many clinical applications related to motion analysis. We propose an innovative approach to compare a markerless tracking system (Kinect V2) with a gold standard marker-based system (Vicon), based on motor control assessment. We quantified kinematic synergies from the tracking data of fifteen participants performing multi-directional upper limb movements. Kinematic synergy analysis decomposes the kinematic data into a reduced set of motor primitives that describe how the central nervous system coordinates motion at spatial and temporal level. Synergies were extracted with the recently released mixed-matrix factorization algorithm. Four synergies were extracted from both marker-based and markerless datasets and synergies were grouped in 6 clusters for each dataset. Cosine similarity in each cluster was ⩾0.60 in both systems, showing good consistency of synergies. Good matching was found between synergies extracted from markerless and from marker-based data, with a cosine similarity between matched synergies ⩾0.60 in 5 out 6 synergies. These results showed that the markerless sensor can be feasible for kinematic synergy analysis for gross movements, as it correctly estimates the number of synergies and in most cases also their spatial and temporal organization.

This study aims to develop an efficient 24-type Tai Chi motion recognition system by using Kinect and mobile edge computing (MEC) technology, to improve the teaching and training effect of Tai Chi. First, by recruiting Tai Chi experts and beginners to conduct experiments, a large number of 24-type Tai Chi movement data are collected to ensure the representativeness and generalization ability of the dataset. Second, in an experimental environment equipped with Kinect devices and MEC nodes, a Tai Chi action recognition model by utilizing Graph Convolutional Network with attention mechanism is constructed. The experimental results reveal that the proposed algorithm achieves 95.17% accuracy in action recognition, markedly higher than the prediction accuracy of the Convolutional Neural Network algorithm. In terms of data transmission, the proposed algorithm’s average delay is stable at 0.53 seconds when the data volume is 6Mb, and the packet loss rate is maintained below 7.7%. The conclusion shows that the proposed algorithm not only has excellent performance in recognition accuracy and real-time performance but also has high stability and reliability in network data transmission performance, which provides effective technical support for Tai Chi’s digital teaching and training.

Objective: Low-cost, portable RGB-D cameras with integrated motion tracking functionality enable easy-to-use 3D motion analysis without requiring expensive facilities and specialized personnel. However, the accuracy of existing systems is insufficient for most clinical applications, particularly when applied to children. In previous work, we developed an RGB-D camera-based motion tracking method and showed that it accurately captures body joint positions of children and young adults in 3D. In this study, the validity and accuracy of clinically relevant motion parameters that were computed from kinematics of our motion tracking method are evaluated in children and young adults. Methods: Twenty-three typically developing children and healthy young adults (5-29 years, 110–189 cm) performed five movement tasks while being recorded simultaneously with a marker-based Vicon system and an Azure Kinect RGB-D camera. Motion parameters were computed from the extracted kinematics of both methods: time series measurements, i.e., measurements over time, peak measurements, i.e., measurements at a single time instant, and movement smoothness. The agreement of these parameter values was evaluated using Pearson’s correlation coefficients r for time series data, and mean absolute error (MAE) and Bland-Altman plots with limits of agreement for peak measurements and smoothness. Results: Time series measurements showed strong to excellent correlations (r-values between 0.8 and 1.0), MAE for angles ranged from 1.5 to 5 degrees and for smoothness parameters (SPARC) from 0.02-0.09, while MAE for distance-related parameters ranged from 9 to 15 mm. Conclusion: Extracted motion parameters are valid and accurate for various movement tasks in children and young adults, demonstrating the suitability of our tracking method for clinical motion analysis. Clinical Impact: The low-cost portable hardware in combination with our tracking method enables motion analysis outside of specialized facilities while providing measurements that are close to those of the clinical gold-standard.

Human activity recognition is known as the backbone of the development of interactive systems, such as computer games. This process is usually performed by either vision-based or depth sensors. So far, various solutions have been developed for this purpose; however, all the challenges of this process have not been completely resolved. In this paper, a solution based on pattern recognition has been developed for labeling and scoring physical exercises performed in front of the Kinect sensor. Extracting the features from human skeletal joints and then generating relative descriptors among them is the first step of our method. This has led to quantification of the meaningful relationships between different parts of the skeletal joints during exercise performance. In this method, the discriminating descriptors of each exercise motion are used to identify the adaptive kernels of the Constrained Energy Minimization method as a target detector operator. The results indicated an accuracy of 95.9% in the labeling process of physical exercise motions. Scoring the exercise motions was the second step after the labeling process, in which a geometric method was used to interpolate numerical quantities extracted from descriptor vectors to transform into semantic scores. The results demonstrated the scoring process coincided with the scores derived by the sports coach by a 99.5 grade in the R2 index.

Effects of bed material and downstream flow depth on bed evolution in a right-angled channel confluence were studied through laboratory experiments. Velocity field, water surface elevation, and bed profile were measured intermittently until the flow attained equilibrium. Sediment discharges at various locations of the confluence were estimated from bed levels. Six prominent bedforms were identified, and their characteristics were quantified. The sediment discharges from different channels were initially high but decreased to approximately zero as the flow attained equilibrium. Downstream flow depth influenced overall sediment transport in the system, including the main features of bed morphology. Overall erosion in the confluence reduced as the particle size of the bed material increased. In addition, the maximum scour depth occurred at the confluence edge, and, due to the sharp corner of the confluence, its location did not change with time. Results from the present experimental study can help validate numerical models and assist in the design of a right-angled confluence. This study is related to the hydraulics of man-made confluences with mobile beds and can be useful in the following practical applications. More accurate numerical models to predict maximum scour, bed morphodynamics, and sediment transport in man-made confluences can be developed using the detailed bed profile measurements at different time instants presented in this study. Note that available numerical models use sediment discharge formulas (e.g., Meyer-Peter-Muller, Van Rijn), which are for sediment flows in straight channels. The interplay between the three-dimensional flow field, developed bed shear stress, and sediment discharge can be explored. In addition, A man-made confluence with the mobile bed can be designed with findings from the present study. The bedforms corresponding to equilibrium flow depend on discharge ratio, confluence angle, flow conditions at the exit channel, and bed material characteristics. The major bedforms can be incorporated into the confluence geometry. [ABSTRACT FROM AUTHOR]

Body composition assessment using instruments such as dual X-ray densitometry (DXA) can be complex and their use is often limited to research. This cross-sectional study aimed to develop and validate a densitometric method for fat mass (FM) estimation using 3D cameras. Using two such cameras, stereographic images, and a mesh reconstruction algorithm, 3D models were obtained. The FM estimations were compared using DXA as a reference. In total, 28 adults, with a mean BMI of 24.5 (±3.7) kg/m2 and mean FM (by DXA) of 19.6 (±5.8) kg, were enrolled. The intraclass correlation coefficient (ICC) for body volume (BV) was 0.98–0.99 (95% CI, 0.97–0.99) for intra-observer and 0.98 (95% CI, 0.96–0.99) for inter-observer reliability. The coefficient of variation for kinetic BV was 0.20 and the mean difference (bias) for BV (liter) between Bod Pod and Kinect was 0.16 (95% CI, −1.2 to 1.6), while the limits of agreement (LoA) were 7.1 to −7.5 L. The mean bias for FM (kg) between DXA and Kinect was −0.29 (95% CI, −2.7 to 2.1), and the LoA was 12.1 to −12.7 kg. The adjusted R2 obtained using an FM regression model was 0.86. The measurements of this 3D camera-based system aligned with the reference measurements, showing the system's feasibility as a simpler, more economical screening tool than current systems. [ABSTRACT FROM AUTHOR]

Falls are a leading cause of unintentional injury and death and pose a significant risk to individuals living alone, particularly older adults. The World Health Organization reports that approximately 684000 fatal falls occur annually, and that timely fall detection is essential to enable prompt treatment and mitigate harm. To address this challenge, we propose a novel inhome fall detection system that uses the Kinect V2 sensor instead of a wearable device. The system can promptly alert a person's caregivers or family members when it detects a fall, enabling timely assistance. The Kinect V2 sensor captures depth frames and performs skeleton tracking, and body parts in single-depth image pixels are identified using a random forest classifier. v-disparity images, the Hough transform, and the random sample consensus algorithm are used to identify the floor in depth images, and falls are predicted by analyzing the vertical accelerations of 10 joints in the human skeleton. The system can issue an alarm up to 0.5 s before a fall, enabling preventive measures to be taken. In experiments, the proposed system was effective in accurately predicting falls and hence, has potential to improve fall risk management in caregiving environments. [ABSTRACT FROM AUTHOR]

Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder known for its significant heterogeneity and varied symptom presentation. Describing the different subtypes as predominantly inattentive (ADHD–I), combined (ADHD–C), and hyperactive–impulsive (ADHD–H) relies primarily on clinical observations, which can be subjective. To address the need for more objective diagnostic methods, this pilot study implemented a Microsoft Kinect-based Stroop Color–Word Test (KSWCT) with the objective of investigating the potential differences in executive function and motor control between different subtypes in a group of children and adolescents with ADHD. A series of linear mixture modeling were used to encompass the performance accuracy, reaction times, and extraneous movements during the tests. Our findings suggested that age plays a critical role, and older subjects showed improvements in KSWCT performance; however, no significant divergence in activity level between the subtypes (ADHD–I and ADHD–H/C) was established. Patients with ADHD–H/C showed tendencies toward deficits in motor planning and executive control, exhibited by shorter reaction times for incorrect responses and more difficulty suppressing erroneous responses. This study provides preliminary evidence of unique executive characteristics among ADHD subtypes, advances our understanding of the heterogeneity of the disorder, and lays the foundation for the development of refined and objective diagnostic tools for ADHD. [ABSTRACT FROM AUTHOR]

The capturing and development of a real time 3D face recognition system is always a big challenge and attracts huge popularity nowadays. This is because, in most security systems employing face recognition, it is always crucial to capture live images and recognize them. This should save time and cost as well, and incredibly is acceptable across all areas of security concerns. This paper deals with the development of an entirely new 3D face recognition system which was developed in Jadavpur University with six subjects captured fully in real time. The main problem which we have tried to address in the present work is that, how various issues like camera calibration, alignment of the camera, distance of the camera from the subject affects the facial recognition rate. We will also analyze how facial registration helps to increase the recognition rates which have been imposed by the above factors. The problem is relevant, because, it's always a challenge to capture and correctly predict a real time system. The necessary problems needed for setting up a real time system is always a matter to be investigated by the researchers in face recognition domain. The main overview of our present proposed method is to capture the subjects in real-time taking into consideration all the calibration issues like camera alignment, distance of camera from the subject, and several other factors. We have also discussed how these factors affect the recognition rate. Once captured, we have tried to justify the varying recognition rate of the subjects due to the changes in calibration, alignment and other issues. Now, in order to improve the performances of the subjects, we have proposed a new 3D face registration algorithm termed as FaRegAvFM8, which was tested on the subjects from our database acquired in real-time, as well as on Frav3D and GavabDB databases. Our system attained a recognition rate of 95.83% after registration on frontal subjects using Haar wavelet as the feature extraction method, which depicts the robustness of the present system. Not only that, we have improved our recognition rate up to 96% using the Deep Convolutional Neural Network (DCNN). [ABSTRACT FROM AUTHOR]

The three-dimensional (3D) geometrical information that depth maps contain is useful in many applications such as 3D reconstruction or simultaneous localization and mapping (SLAM). Kinect is widely used in depth image acquisition due to its low cost and good real-time performance. However, the quality of depth images obtained by Kinect is influenced by holes which make depth image inadequate for further applications. To suppress the influence of holes on a subsequent application, a fixing algorithm of Kinect depth image based on non-local means (NLM) is proposed in this paper. The holes in depth image are filled using the weights which are calculated on the corresponding gray image by distance factor and value consistent factor. And the experiment results demonstrate that the proposed method achieves good performance in both evaluation in metrics and subjectively visual effect. This research provides a solution idea for depth image fixing algorithm with low complexity. [ABSTRACT FROM AUTHOR]

Everyone aspires to live a healthy life, but many will inevitably experience some form of disease, illness, or accident that results in disability at some point. Rehabilitation plays a crucial role in helping individuals recover from these disabilities and return to their daily activities. Traditional rehabilitation methods are often expensive, are inefficient, and lead to slow progress for patients. However, in this era of technology, various sensor-based automatic rehabilitation is also possible. A Kinect sensor is a skeletal tracking device that captures human motions and gestures. It can provide feedback to the users, allowing them to better understand their progress and adjust their movements accordingly. In this study, stroke-based rehabilitation is presented along with the Toronto Rehab Stroke Pose Dataset (TRSP). Pre-processing of the raw dataset was performed using various features, and several state-of-the-art classifiers were applied to evaluate the data provided by the Kinect sensor. Among the various classifiers, eXtreme Gradient Boosing (XGB) attained the maximum accuracy of 92% for the TRSP dataset. Furthermore, hyperparameters of the XGB have been optimized using a metaheuristic gray wolf optimizer for better performance.

Aiming at the current lack of innovation in Chinese ceramic culture, this paper proposes an inheritance and innovation path combining ceramic digitization technology and ceramic virtual display technology. Based on the three-dimensional morphology reconstruction technology method, using a three-dimensional scanner in different groups of lamps under the scanning pottery plastic to obtain point cloud data, using an iterative proximity point algorithm to achieve point cloud data alignment, and obtaining a complete three-dimensional model of pottery plastic. Then, we propose the ceramic virtual display technology based on Kinect, using Kinect to capture human motion data, designing somatosensory interaction subjects, expanding Kinect human-computer interaction actions, and providing an interactive and experiential ceramic culture display. Using ceramic digital technology to build three-dimensional digital modeling of the five famous kilns in China, the simulation effect evaluation score is greater than 80 points, of which the Gao kiln has the highest average score of 90 points. The use of ceramic virtual display technology to construct a virtual ceramic museum of the Maritime Silk Road demonstrates good system performance. In each satisfaction evaluation factor, the online visitors exhibit a general, satisfied, or very satisfied attitude, with a proportion exceeding 60%.

Augmented reality (AR) technology has rapidly advanced across various domains, propelled by its robust interactive immersion and the seamless integration of real and virtual environments. However, its exploration and deployment in theatrical contexts remain limited. This study leverages the Kinect system to capture images during costume performances, employing algorithms for dynamic frame difference merging and human-computer interaction to detect performers’ body movements. Building on this, the study constructs a visually innovative stage for costume performances that enhances the interactive experience for the audience. Additionally, a multimodal emotion analysis model is utilized to assess audience emotions, demonstrating significantly higher accuracy and F1 scores compared to other emotion analysis models. This model effectively integrates speech, expression, and action, surpassing the performance of unimodal analyses in emotion recognition. Furthermore, the audience's experiential perception of stage lighting effects notably exceeds expectations (P=0.013 < 0.05), underscoring an enhanced interaction experience. This research substantiates the transformative potential of AR technology in stage design, offering audiences a more innovative visual and interactive experience, and serves as a valuable reference for future applications in this field.

The integration of digital media art into the public art design of the smart city can enhance people’s participation and sense of identity in the art design of public space and maximize the attractiveness and cultural connotation of the city itself. Based on this, this paper builds a fusion technology framework mainly based on digital projection technology and digital interaction technology, proposes an interactive projection technology integrating digital projection and Kinect, and combines the dynamic scene image splicing and alignment fusion technology to realize the interaction between the user and the digital scene. In the case analysis of Qianjiang New City, the number of visitors reached 31.55 million in 2022. The evaluation score of landscape facilities in the public evaluation of architectural landscape is under 3.5. The overall evaluation grade is only “general,” and the evaluation mean value of the degree of intelligence of landscape facilities is only 3.089. The comprehensive evaluation satisfaction of the digital landscape includes multimedia landscape, architectural landscape, and landscape facilities, and the average value of the evaluation of the degree of intelligence of the landscape facilities is only 3.089. The satisfaction value of the multimedia landscape, landscape of buildings and structures, and water landscape in the evaluation of satisfaction is 4.36, 4.159, and 4.027, respectively, which reaches the level of “good” satisfaction, while the satisfaction value of landscape facilities is 3.479, which is less than 3.5. The satisfaction level only reaches the level of “general.” The satisfaction value of landscape facilities is 3.479, which is less than 3.5, and the satisfaction level is only “fair.”

Face orientation recognition is a crucial component of human behavior analysis. Challenges such as face occlusion and varying lighting conditions frequently degrade recognition accuracy. This study introduces an innovative approach using the Kinect sensor’s skeleton tracking technology to capture face orientation images across different lighting scenarios. To enhance image quality, techniques such as light attenuation and compensation are applied. Furthermore, multiple feature vectors based on face orientation are extracted and fused, followed by the optimization of the Extreme Learning Machine (ELM) model through the Archimedean Optimization Algorithm (AOA). The developed AOA-ELM model for face orientation recognition is compared with traditional models, including Back Propagation (BP), Radial Basis Function (RBF), standard ELM, and Genetic Algorithm enhanced ELM (GA-ELM). The AOA-ELM model exhibits superior performance in estimating face orientation amidst lighting disturbances, achieving an accuracy exceeding 95% in varied lighting conditions, thereby outperforming other classification models.

This paper combines Kinect and convolutional neural networks to construct a dance movement recognition technology based on 3D CNNs. Applying dance movement recognition technology to dance teaching builds a new mode of training dance talents in colleges and universities. The role of Kinect in recognizing dance movements is explored from three aspects: real-time motion capture, human skeleton tracking, and information input. The dance movement image is computerized by calculating the depth of the points using stereo analysis. To analyze dance movements and classify them by features, a convolutional neural network is combined. Based on two-dimensional convolutional neural networks, three-dimensional convolutional neural networks have been constructed, which improve the comprehensiveness of dance movement information. By combining dance movement recognition technology with dance talent cultivation, we analyze the students’ professional dance ability and the teaching effect under the new talent cultivation mode. The results show that the teaching effect of the dance talent cultivation mode combined with the movement recognition technology is better, and the percentage of students dance movements reaching the standard in a movement completion is 0.95. The professional ability of dance talent is 0.8 percent.

With the increasing prosperity of the cultural market, the dance culture industry is also developing rapidly. In the face of opportunities and challenges brought by dance industrialization, dance professionals trained by the traditional and single-skill training system of art colleges and universities expose the contradiction between supply and demand. In this paper, the SAA-training model is taken as the calculation tool, and the OBE and Kinect motion capture technology are combined to explore the construction and practice of the training system for dance professionals. The research shows that the training system of dance professionals can finally obtain three types of dance professionals. The first type is compound dance talents, accounting for about 33%. The second category consists of dancers with rich market experience, accounting for up to 46 percent. The third category is dance talent with innovative consciousness, limited by the existing training mode, its proportion is only 21 percent.

In this paper, a static gesture classification model based on a convolutional neural network is proposed to achieve the goal of efficient interaction control and fast response time for gesture recognition, and hand feature extraction is achieved by using alternating convolution and pooling. In the convolutional neural network, image data is obtained by preprocessing operations such as adaptive thresholding and functions such as Gaussian filtering to provide data for the convolutional neural network structure to build the recognition model, which is used to quantify the edge characteristics of the image and construct the image edges with gradient amplitude to build the static gesture recognition algorithm for Kinect. In the accuracy rate for gesture interaction, the average recognition rate for slow movement is 96.5%, while when the interactive gesture is recognized at an accelerated speed, its recognition rate slips to 93.1%, and the total average recognition rate reaches 95.2%.

Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder known for its significant heterogeneity and varied symptom presentation. Describing the different subtypes as predominantly inattentive (ADHD–I), combined (ADHD–C), and hyperactive–impulsive (ADHD–H) relies primarily on clinical observations, which can be subjective. To address the need for more objective diagnostic methods, this pilot study implemented a Microsoft Kinect-based Stroop Color–Word Test (KSWCT) with the objective of investigating the potential differences in executive function and motor control between different subtypes in a group of children and adolescents with ADHD. A series of linear mixture modeling were used to encompass the performance accuracy, reaction times, and extraneous movements during the tests. Our findings suggested that age plays a critical role, and older subjects showed improvements in KSWCT performance; however, no significant divergence in activity level between the subtypes (ADHD–I and ADHD–H/C) was established. Patients with ADHD–H/C showed tendencies toward deficits in motor planning and executive control, exhibited by shorter reaction times for incorrect responses and more difficulty suppressing erroneous responses. This study provides preliminary evidence of unique executive characteristics among ADHD subtypes, advances our understanding of the heterogeneity of the disorder, and lays the foundation for the development of refined and objective diagnostic tools for ADHD.

In recent years, the incidence rate of children with autism has shown a significant upward trend. Rehabilitation training is an important part of recovery or improvement in autism children. However, during autism rehabilitation training, the methods that can visually reflect and objectively evaluate its effects are seldom considered. Therefore, this study aimed to objectively evaluate the rehabilitation impact of visual-motor skills training in children with autism via quantitative measures. In this study, vision sensors and functional near-infrared spectroscopy were used to monitor and analyze visual motor training task of 20 autism children. These children were divided into high- and low-score groups according to the autism behavior checklist (ABC). Results showed significant differences between the high- and low-score groups in the brain regions of the left and right temporal lobe, right motor cortex, and left occipital lobe; the difference in functional connectivity was greatest when the left hand was moving at the green light (p < 0.05). The differences in speed, acceleration, and angle between the high- and low-score groups were mainly reflected in left-hand movement. Moreover, analysis of multimodal data showed that visual motor training had a positive effect on brain activation and functional connectivity, and increasing the frequency of left-hand training and using more green light were beneficial to the improvement of brain function. These findings can be used as basis to help optimize rehabilitation programs and improve rehabilitation effectiveness., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 International Brain Research Organization (IBRO). Published by Elsevier Inc. All rights reserved.)

Objective: Low-cost, portable RGB-D cameras with integrated motion tracking functionality enable easy-to-use 3D motion analysis without requiring expensive facilities and specialized personnel. However, the accuracy of existing systems is insufficient for most clinical applications, particularly when applied to children. In previous work, we developed an RGB-D camera-based motion tracking method and showed that it accurately captures body joint positions of children and young adults in 3D. In this study, the validity and accuracy of clinically relevant motion parameters that were computed from kinematics of our motion tracking method are evaluated in children and young adults., Methods: Twenty-three typically developing children and healthy young adults (5-29 years, 110-189 cm) performed five movement tasks while being recorded simultaneously with a marker-based Vicon system and an Azure Kinect RGB-D camera. Motion parameters were computed from the extracted kinematics of both methods: time series measurements, i.e., measurements over time, peak measurements, i.e., measurements at a single time instant, and movement smoothness. The agreement of these parameter values was evaluated using Pearson's correlation coefficients r for time series data, and mean absolute error (MAE) and Bland-Altman plots with limits of agreement for peak measurements and smoothness., Results: Time series measurements showed strong to excellent correlations (r-values between 0.8 and 1.0), MAE for angles ranged from 1.5 to 5 degrees and for smoothness parameters (SPARC) from 0.02-0.09, while MAE for distance-related parameters ranged from 9 to 15 mm., Conclusion: Extracted motion parameters are valid and accurate for various movement tasks in children and young adults, demonstrating the suitability of our tracking method for clinical motion analysis., Clinical Impact: The low-cost portable hardware in combination with our tracking method enables motion analysis outside of specialized facilities while providing measurements that are close to those of the clinical gold-standard., (© 2024 The Authors.)

Recently, markerless tracking systems, such as RGB-Depth cameras, have spread to overcome some of the limitations of the gold standard marker-based tracking systems. Although these systems are valuable substitutes for human motion analysis, as they guarantee higher flexibility, faster setup time and lower costs, their tracking accuracy is lower with respect to marker-based systems. Many studies quantified the error made by markerless systems in terms of body segment length estimation, articular angles, and biomechanics, concluding that they are appropriate for many clinical applications related to motion analysis. We propose an innovative approach to compare a markerless tracking system (Kinect V2) with a gold standard marker-based system (Vicon), based on motor control assessment. We quantified kinematic synergies from the tracking data of fifteen participants performing multi-directional upper limb movements. Kinematic synergy analysis decomposes the kinematic data into a reduced set of motor primitives that describe how the central nervous system coordinates motion at spatial and temporal level. Synergies were extracted with the recently released mixed-matrix factorization algorithm. Four synergies were extracted from both marker-based and markerless datasets and synergies were grouped in 6 clusters for each dataset. Cosine similarity in each cluster was ⩾0.60 in both systems, showing good consistency of synergies. Good matching was found between synergies extracted from markerless and from marker-based data, with a cosine similarity between matched synergies ⩾0.60 in 5 out 6 synergies. These results showed that the markerless sensor can be feasible for kinematic synergy analysis for gross movements, as it correctly estimates the number of synergies and in most cases also their spatial and temporal organization., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2024 The Author(s).)