Skeleton Model Research Articles

Correlation dimension and entropy in the assessment of sex differences based on human gait data.

It is proved that there are differences between gait performed by females and males, which appear in movements of selected body parts. Despite numerous state-of-the-art studies related to the discriminative analysis of motion capture data, the question of whether measures of signal complexity and uncertainty can extract valuable features for the problem of sex distinction still remains open. It is the subject of the paper. Correlation dimension, as well as approximate and sample entropies, are selected to describe motion data. In the numerical experiments, the collected dataset with 884 samples of 25 females and 30 males was used. The measurements took place in the Human Motion Laboratory (HML), equipped with a highly precise motion capture system. Two variants of data representation were investigated-time series that contain joint rotations of taken skeleton model as well as positions of the markers attached to the human body. Finally, a comparative analysis between the populations of females and males using descriptive statistics, non-parametric estimation, and statistical hypotheses verification was carried out. There are statistically significant sex differences extracted by the taken measures. In general, the movements of lower limbs result in greater values of correlation dimension and entropies for females, while selected upper body parts play a similar role for males. The dissimilarities are mainly observed in hip, ankle, shoulder, and head movements. Correlation dimension and entropy measures provide robust and explainable features of motion capture data with a valuable description of the human locomotion system. Thus, beyond the importance of discovered differences between females and males, their interpretation and understanding are also known.

Fast Detection Algorithm for Baggage Pallet Based on Skeleton Model

Fast Detection Algorithm for Baggage Pallet Based on Skeleton Model

Micro-expression recognition based on multi-scale 3D residual convolutional neural network.

In demanding application scenarios such as clinical psychotherapy and criminal interrogation, the accurate recognition of micro-expressions is of utmost importance but poses significant challenges. One of the main difficulties lies in effectively capturing weak and fleeting facial features and improving recognition performance. To address this fundamental issue, this paper proposed a novel architecture based on a multi-scale 3D residual convolutional neural network. The algorithm leveraged a deep 3D-ResNet50 as the skeleton model and utilized the micro-expression optical flow feature map as the input for the network model. Drawing upon the complex spatial and temporal features inherent in micro-expressions, the network incorporated multi-scale convolutional modules of varying sizes to integrate both global and local information. Furthermore, an attention mechanism feature fusion module was introduced to enhance the model's contextual awareness. Finally, to optimize the model's prediction of the optimal solution, a discriminative network structure with multiple output channels was constructed. The algorithm's performance was evaluated using the public datasets SMIC, SAMM, and CASME Ⅱ. The experimental results demonstrated that the proposed algorithm achieves recognition accuracies of 74.6, 84.77 and 91.35% on these datasets, respectively. This substantial improvement in efficiency compared to existing mainstream methods for extracting micro-expression subtle features effectively enhanced micro-expression recognition performance and increased the accuracy of high-precision micro-expression recognition. Consequently, this paper served as an important reference for researchers working on high-precision micro-expression recognition.

Research on Dynamic Capture and Pattern Recognition Technology of Pitching Technology in Baseball Sports

Abstract This paper proposes a baseball pitching action recognition algorithm based on a spatiotemporal graph convolutional neural network and constructs an error action correction algorithm on this basis. The dynamic skeleton model ST-GCN is used to combine the positional information of human movement with the temporal dynamic information. The action contour sequence is extracted to determine the funding for the erroneous action. Finally, the machine learning method is used to realize the adaptive corrective analysis of the erroneous action. Example analysis shows that the action correction algorithm proposed in this paper improves the recognition accuracy by 20.78%, 16.67%, and 9.11%, 9.73% in the two datasets, and the pitching accuracy of the experimental group is 12.5% higher than that of the control group, and the standardized degree score of the pitching technical action is 1.1 points higher than that of the control group. Therefore, the practical effectiveness of the pitching action identification and correction method in this paper has been effectively verified.

Application of 3D recognition algorithm based on spatio-temporal graph convolutional network in basketball pose estimation

In recent years, human motion recognition in computer vision has become a hot research direction in this field. Based on 2D human motion recognition technology, real-time extraction of motion features from 2D planes is used to recognize human movements. This method can only learn the position contour and color information of the image. It cannot directly reflect the motion situation, which results in low recognition accuracy and efficiency. In response to this issue, this study proposes a combination method of motion recognition and 3D pose estimation to recognize and classify basketball movements. First, the 2D skeleton model is obtained by extracting the feature information in the video action, which is converted into a 3D model, and the model is replaced by the time-space convolutional network to establish a human action recognition model. The experiment showed that when the number of iterations reached 6, the accuracy of the spatio-temporal graph convolutional network algorithm model reached 92%. Comparing the accuracy rates of different algorithm models, the average accuracy rates of convolutional neural network, long short memory network, graph convolution, long short model of action recognition and graph convolution model of action recognition were 61.6%, 65.4%, 72.5%, 76.8% and 90.3% respectively. The results show that the proposed 3D recognition algorithm can accurately recognize different basketball movements. This study can provide reference for basketball coaches and athletes in basketball training.

Research on driver posture recognition and prediction in vehicle

The statistics show that more than 90% of traffic accidents are caused by the individual factors of the driver. Therefore, real-time monitoring of the current driver’s posture in the car and further predicting of the future driving posture are of great significance for designing the Adaptive Restraint System (ARS) to improve the safety protection of drivers. In this study, a real-time vehicle driver posture recognition and prediction method is proposed. The traditional Kinect-OpenPose algorithm, which has the problems such as lack of depth value, key point pulsation and measurement error of Kinect, was optimized. The OpenPose network structure and skeleton model are modified to improve the algorithm. In this study, a key point motion trajectory prediction network based on Long Short Term Memory (LSTM) is developed. The hyperparameter optimization method is combined with multi-layer grid search to obtain the best prediction effect in braking and steering conditions, and the driver’s future posture can be obtained by predicting the subsequent movement trajectory of key points. It has been verified that the method in this study has high accuracy, high robustness and low cost in key point recognition and prediction. And the improved processing speed of the algorithm did not affect the accuracy of driving posture recognition and prediction. This method predicted the driver’s posture in future of the collision, which can be used to guide the development of controlling system of the ARS.

Research on the Innovation of Physical Education and Physical Education Teaching Based on Big Data Analysis

Abstract In this paper, Openpose is used to process the sports teaching video to get the coordinates data of the human body joint point positions in each frame of the video, and Kalman filter data fusion is used to establish the human skeleton model. According to the results of the division of the five major parts of the human body, after establishing the limb vectors of the human body’s torso, left arm, right arm, left leg, and right leg in three-dimensional spatial coordinates, the distances between the joints of the five human body skeletons based on the DTW posture matching algorithm were used to extract the characteristics of the sports error technical movements. From the demand of sports digital teaching, the design and implementation of sports basic movement teaching evaluation system based on the DTW posture matching algorithm, and the research and analysis of sports teaching under the background of big data. The results show that the IoU values of batting action localization in 6 segments of physical education teaching are 85.6%, 91.6%, 77.7%, 75.1%, 87.4% and 77.7%, respectively, and the average reach 82.5%, i.e., it shows that the research on action localization and recognition based on the DTW posture matching algorithm has a good performance. In the assessment of movement standardization in physical education, the maximum moment of stretching angle corresponds to the moment of hitting the ball, and its value reaches 3.79, i.e., it reflects that the evaluation system of physical education basic movement teaching can accurately determine whether the students’ movements are accurate or not, and make timely feedbacks to carry out the corrections of physical education movements. This study has the potential to enhance students’ interest and performance in sports and contribute to the advancement of digital sports teaching.

Research on Deep Learning-based Group Recognition

The task of group recognition is a process of automatically recognizing and analyzing individuals or groups from a large population using computer vision and machine learning techniques. Currently, the main methods for group recognition using deep networks can be divided into four categories: methods based on convolutional neural networks (CNN), methods based on recurrent neural networks (RNN), combined CNN-RNN-based methods, as well as pose analysis methods. This article specifically analyzes these four categories of methods: CNN-based, RNN-based, and pose analysis, and summarizes the characteristics and functions of iterative deep learning models, graph convolutional network models, long short-term memory models, pose models, and skeleton models. In addition, a detailed introduction to commonly used datasets in group recognition is provided, and the main evaluation metrics are analyzed. Finally, important research directions in group recognition are discussed, followed by a conclusion of the entire article.

Seq2seq model for human action recognition based on skeleton and two-layer bidirectional LSTM

Human action recognition (HAR) plays an important role in social interaction in various fields. This study proposes a light-weight skeleton and two-layer bidirectional LSTM-based Seq2Seq model (SB2_Seq2Seq) for HAR to trade off recognition accuracy, users’ privacy and computer resource usage. An experiment was conducted to compare the proposed SB2_Seq2Seq with other skeleton-based Seq2Seq models and non-skeleton RGB video frame-based LSTM, CNN and seq2seq models. The UCF50 dataset was used for model evaluation, where 60%, 20% and 20% for model training, validation and testing, respectively. The experimental results show that the proposed model achieves 93.54% accuracy with 0.0214 Mean Square Error (MSE), suggesting that the proposed model outperforms all the other models. Besides, it also shows that the proposed model achieves state-of-the-art accuracy compared with state-of-the-arts methods in literature.

Biosensor-Based Multimodal Deep Human Locomotion Decoding via Internet of Healthcare Things.

Multiple Internet of Healthcare Things (IoHT)-based devices have been utilized as sensing methodologies for human locomotion decoding to aid in applications related to e-healthcare. Different measurement conditions affect the daily routine monitoring, including the sensor type, wearing style, data retrieval method, and processing model. Currently, several models are present in this domain that include a variety of techniques for pre-processing, descriptor extraction, and reduction, along with the classification of data captured from multiple sensors. However, such models consisting of multiple subject-based data using different techniques may degrade the accuracy rate of locomotion decoding. Therefore, this study proposes a deep neural network model that not only applies the state-of-the-art Quaternion-based filtration technique for motion and ambient data along with background subtraction and skeleton modeling for video-based data, but also learns important descriptors from novel graph-based representations and Gaussian Markov random-field mechanisms. Due to the non-linear nature of data, these descriptors are further utilized to extract the codebook via the Gaussian mixture regression model. Furthermore, the codebook is provided to the recurrent neural network to classify the activities for the locomotion-decoding system. We show the validity of the proposed model across two publicly available data sampling strategies, namely, the HWU-USP and LARa datasets. The proposed model is significantly improved over previous systems, as it achieved 82.22% and 82.50% for the HWU-USP and LARa datasets, respectively. The proposed IoHT-based locomotion-decoding model is useful for unobtrusive human activity recognition over extended periods in e-healthcare facilities.

Multiscale topology optimization of gradient lattice structure based on volume parametric modeling

In this study, a volume parametric modeling method of lattice structure is proposed, and an efficient multiscale topology optimization framework is realized based on isogeometric analysis (IGA) to construct the gradient lattice structure. The skeleton model is constructed, which can accurately describe the topology structure and improve data utilization and computational efficiency. Based on the skeleton model, a uniform volume parameter lattice structure with an arbitrary topology of multiple types of unit cells is constructed, which is suitable for IGA. Moreover, multiscale topology optimization based on IGA is realized to construct the gradient lattice structure. The same data model is used in modeling, analysis, and optimization, which can accurately represent the geometric shape without discretization errors. At the same time, the multiscale topology optimization iteration is realized by adjusting the density of control points. The optimized model can be directly analyzed and re-optimized, thus realizing the integrated design of lattice structure modeling, simulation, and optimization. The effectiveness and robustness of the algorithm are verified by several mechanical parts and freeform models. These examples show that the gradient lattice structure has higher strength and better stress distribution than the uniform lattice structure under the same boundary conditions.

EPCS: Endpoint-based part-aware curve skeleton extraction for low-quality point clouds

The curve skeleton is an important shape descriptor which has been utilized in various applications in computer graphics, machine vision, and artificial intelligence. In this study, the endpoint-based part-aware curve skeleton (EPCS) extraction method for low-quality point clouds is proposed. The novel random center shift (RCS) method is first proposed for detecting the endpoints on point clouds. The endpoints are used as the initial seed points for dividing each part into layers, and then the skeletal points are obtained by computing the center points of the oriented bounding box (OBB) of the layers. Subsequently, the skeletal points are connected, thus forming the branches. Furthermore, the multi-vector momentum-driven (MVMD) method is also proposed for locating the junction points which connect the branches. Due to the shape differences between different parts on point clouds, the global topology of the skeleton is finally optimized by removing the redundant junction points, re-connecting some branches using the proposed MVMD method, and applying an interpolation method based on the splitting operator. Consequently, a complete and smooth curve skeleton is achieved. The proposed EPCS method is compared with several state-of-the-art methods, and the experimental results verify its robustness and effectiveness. Furthermore, the skeleton extraction and model segmentation results on challenging point clouds of broken Terracotta also highlight the utility of the proposed method.

Synthesis of lithium slag-based ANA zeolite and molecular dynamics study of carbon dioxide adsorption performance

In this paper, lithium slag was used as raw material to prepare zeolite framework with porous structure by alkali activation and hydrothermal synthesis. The alkali activation of raw materials and the phase analysis of zeolite products were carried out by XRD characterization methods, and the effects of different alkali fusion conditions and hydrothermal conditions on zeolite synthesis were investigated. At the same time, the effects of different temperatures on the adsorption capacity and adsorption heat of carbon dioxide adsorbed by zeolite skeleton model under normal pressure were studied by using Materials studio software. The results showed that the optimum alkali fusion conditions were: 500 °C, 6 h; the best hydrothermal condition is 220 °C, 6 h. The average adsorption capacity of cristobalite for carbon dioxide is 2 mmol/g, and the adsorption heat is 11.98 kJ/mol.

Three-dimensional digital technology-assisted precise tumor resection and reconstruction of the femoral trochanter and postoperative functional recovery: a retrospective study

BackgroundThe trochanter of the femur is a common site for bone tumors. However, locating the specific boundary of bone tumor infiltration and determining the surgical method can be challenging. The objective of this study was to review the diagnosis, treatment, and surgical outcomes of patients with tumors or tumor-like changes in the femoral trochanter after computer-assisted precise tumor resection and hip-preserving reconstruction of the trochanter. MethodsFrom January 2005 to September 2020, 11 patients with trochanteric tumors (aged: 18–53 years; six males and five females) were treated in Guangzhou First People's Hospital. The cases included aneurysmal bone cyst (n = 1), giant cell tumor of bone (n = 2), fibrous histiocytoma of bone (n = 1), endochondroma (n = 1), and fibrous dysplasia of bone (n = 6). For patients with trochanteric tumors, computed tomography and magnetic resonance imaging scanning were performed before operation to obtain two-dimensional image data of the lesion. A three-dimensional digital model of bilateral lower limbs was reconstructed by computer technology, the boundary of tumor growth was determined by computer simulation, the process of tumor resection and reconstruction was simulated, and the personalized guide template was designed. During the operation, the personalized guide plate guided the precise resection of the tumor, and the allogeneic bone was trimmed to match the shape of the bone defect. ResultsAll 11 patients underwent accurate resection of the tumor or tumor-like lesion and reconstruction of the hip. In eight cases, the lesion was confined to the trochanter, which was fixed with large segment allogeneic bone, autologous iliac bone, and proximal femoral anatomic plate. In three cases, allogeneic bone, autologous iliac bone, and femoral reconstruction nail were used to fix the tumor under the trochanter. Postoperative X-ray examination showed that the repair and reconstruction of the bone defect was effective, and callus bridging between the allogenic bone and autogenous bone was observed 6 months after operation. All patients recovered their walking function 3–6 months after operation. The duration of the follow-up period ranged from 6 months to 6 years. A patient experienced recurrence of endochondroma; pathological examination revealed chondrocytic sarcoma. The remaining 10 patients were treated with segmental resection and reconstruction. The operation time ranged 2.5–4.5 h (average: 3.2 h). Intraoperative blood loss ranged from 300 to 500 ml (average: 368 ml). The local recurrence rate was 9.1%, and the overall survival rate was 100%. The average Musculoskeletal Tumor Society score was 27 (excellent and good for eight and three patients, respectively). ConclusionsThree-dimensional computer skeleton modeling and simulation-assisted resection and reconstruction of femoral trochanteric tumor is a new surgical technique, which might markedly improve the surgical effect, shorten the surgical time, increase the overall survival rate of patients with tumors, reduce the local recurrence rate, assist in the digitization and programming of femoral trochanteric tumor surgery, and improve surgical accuracy.

Deep Learning in Sign Language Recognition: A Hybrid Approach for the Recognition of Static and Dynamic Signs

A speech impairment limits a person’s capacity for oral and auditory communication. A great improvement in communication between the deaf and the general public would be represented by a real-time sign language detector. This work proposes a deep learning-based algorithm that can identify words from a person’s gestures and detect them. There have been many studies on this topic, but the development of static and dynamic sign language recognition models is still a challenging area of research. The difficulty is in obtaining an appropriate model that addresses the challenges of continuous signs that are independent of the signer. Different signers’ speeds, durations, and many other factors make it challenging to create a model with high accuracy and continuity. For the accurate and effective recognition of signs, this study uses two different deep learning-based approaches. We create a real-time American Sign Language detector using the skeleton model, which reliably categorizes continuous signs in sign language in most cases using a deep learning approach. In the second deep learning approach, we create a sign language detector for static signs using YOLOv6. This application is very helpful for sign language users and learners to practice sign language in real time. After training both algorithms separately for static and continuous signs, we create a single algorithm using a hybrid approach. The proposed model, consisting of LSTM with MediaPipe holistic landmarks, achieves around 92% accuracy for different continuous signs, and the YOLOv6 model achieves 96% accuracy over different static signs. Throughout this study, we determine which approach is best for sequential movement detection and for the classification of different signs according to sign language and shows remarkable accuracy in real time.

Mobile Device-Based Video Screening for Infant Head Lag: An Exploratory Study.

Video-based automatic motion analysis has been employed to identify infant motor development delays. To overcome the limitations of lab-recorded images and training datasets, this study aimed to develop an artificial intelligence (AI) model using videos taken by mobile phone to assess infants' motor skills. A total of 270 videos of 41 high-risk infants were taken by parents using a mobile device. Based on the Pull to Sit (PTS) levels from the Hammersmith Motor Evaluation, we set motor skills assessments. The videos included 84 level 0, 106 level 1, and 80 level 3 recordings. We used whole-body pose estimation and three-dimensional transformation with a fuzzy-based approach to develop an AI model. The model was trained with two types of vectors: whole-body skeleton and key points with domain knowledge. The average accuracies of the whole-body skeleton and key point models for level 0 were 77.667% and 88.062%, respectively. The Area Under the ROC curve (AUC) of the whole-body skeleton and key point models for level 3 were 96.049% and 94.333% respectively. An AI model with minimal environmental restrictions can provide a family-centered developmental delay screen and enable the remote monitoring of infants requiring intervention.

Research on the resilience model of a new prefabricated shear wall structure

Based on high-strength bolts, steel plates and other components to achieve the advantages of prefabricated shear wall assembly connection “dry connection” operation advantages, a new type of assembled shear wall with friction energy consumption function was proposed. Using a combination of model tests and numerical simulations, three shear wall specimens with a scaling ratio of 1:1.54 were designed and the corresponding seismic performance tests were carried out. The effects of axial compression ratio and longitudinal reinforcement rate of edge members on the damage mode, hysteresis performance, load carrying capacity, deformation performance, stiffness degradation and energy dissipation capacity of specimens are systematically analyzed. The test results showed that the final failure modes of the three specimens were similar, which was the bending-shear failure from the development of the oblique cracks in the wall to the concrete spalling in the compression zone; The reduction of the axial compression ratio or the reinforcement rate of the longitudinal reinforcement of the edge members will lead to the reduction of the bearing capacity of the shear wall and the increase of the ultimate displacement has a negative impact on the seismic performance of the new assembled shear wall, which is manifested by the weakening of the hysteresis performance and the reduction of the ultimate bearing capacity. Combined with the finite element simulation results, the five-fold skeleton model and restoring force model of the new assembled shear wall were established, and then the calculation formulae of unloading stiffness and loading stiffness of the hysteresis curve were obtained by regression fitting. The restoring force model proposed in this paper is in good agreement with the test results, which can provide a reference for the elastic–plastic seismic response analysis of the new fabricated shear wall structure.

Seismic behavior and simplified hysteretic model of precast concrete wall panels with bolted connections under cyclic loading

Bolted prefabricated concrete wall panel structures are considered to be of great advantage in the current construction industry as they allow efficient recycling or direct replacement of aging components for structural renewal and repair. However, the parts of bolted connectors still have many problems such as complex construction and high positioning accuracy requirements. Furthermore, comparative studies on different bolted connections have been limited to date. This paper proposes a simplified tensile-bolt connection and a shear bolt-steel plate connection for low-rise wall-based structures in earthquake-prone zones, which can adapt to the fast-easy and reliable installation manners on construction sites. Three tension-bolted specimens and three shear-bolted specimens were tested under quasi-static cyclic loading. The seismic performances studied include failure modes, hysteretic behaviors, deformation, energy dissipation capacity, stiffness degradation, and steel strain distributions. The results indicated that both types of bolted specimens meet the seismic safety of low-rise buildings. By analyzing the typical skeleton curve and hysteretic loop of the specimens, based on a linear regression fitting analysis, the characteristic points of the proposed skeleton model curves were proposed which could evaluate the restoring characteristics of the curves well. A simplified restoring force model was proposed which has a high evaluation accuracy. The research results can give references to the further application of bolt-connected fabricated wall panel structures.

High performance fiber-constrained plasticized PVC gel actuators for soft robotics

Plasticized Poly (vinyl chloride) (PVC) gel actuators, a type of electroactive soft actuator, have recently attracted increasing interest in soft robotics. Herein, we report a novel planar PVC gel actuator with flexible electrodes and a fiber-constrained structure for electroactive artificial muscles. PET adhesive fibers were evenly affixed on both sides of the unidirectional pre-stretched PVC gel membranes in one direction, resulting in a human-muscle-like unidirectional contraction and expansion movement. The fibers can constrain the bending motion of the PVC gel membranes, and the unidirectional pre-stretch in the direction of the attached fibers further makes the membrane anisotropic, leading to easy and fast actuation in the direction normal to the pre-stretch. The developed fiber-constrained plasticized PVC gel actuator (FCPPGA) unit exhibited high strain (>40%) and stress (>0.5 MPa) with a fast response time of about 0.1 s under a relatively low driving DC field (<15 V/μm). The simulated data were in good agreement with the experimental results. Furthermore, the designed FCPPGA was successfully used to demonstrate the mimicry of lower jaw motions in a skeleton model.

Action recognition for sports combined training based on wearable sensor technology and SVM prediction

In the field of sports, coaches have mainly relied on observing the performance of athletes on the spot to formulate suitable training plans for athletes, which has extremely high requirements for the professionalism of coaches. Based on the above requirements, this paper designs a sports action recognition system for sports enthusiasts based on the SVM algorithm optimization model, and for the purpose of verifying the applicability of the system to different sports fields, experiments are carried out on basketball actions and race walking actions. The system uses wearable sensors to capture the motion data of the user, and then analyzes and identifies the user's actions through the SVM algorithm optimization model. By standardizing the user's sports combination training under the system algorithm, the user can improve their training efficiency and reduce the risk of injury. To establish the human body motion model, this paper divides the human skeleton model into five motion branches. The rotation freedom constraints and joint rotation angle range limits are added to the model to ensure the accuracy of the motion analysis. Combining the forward kinematics of the robot and the homogeneous coordinate transformation, the human body joint rotation motion model and the human bone position and posture model are established. In the end, the user can standardize the sports combination training under the system algorithm. In this paper, through the research of wearable sensor technology and sports combined training action recognition, and apply it to practical life, it aims to promote its development and application.