Visual Motion Segmentation in Crowd Videos Based on Spatial-Angular Stacked Sparse Autoencoders

Visual motion segmentation(VMS)is an important and key part of many intelligent crowd systems.It can be used to figure out the flow behavior through a crowd and to spot unusual life-threatening incidents like crowd stampedes and crashes,which pose a serious risk to public safety and have resulted in numerous fatalities over the past few decades.Trajectory clustering has become one of the most popular methods in VMS.However,complex data,such as a large number of samples and parameters,makes it difficult for trajectory clustering to work well with accurate motion segmentation results.This study introduces a spatial-angular stacked sparse autoencoder model(SA-SSAE)with l2-regularization and softmax,a powerful deep learning method for visual motion segmentation to cluster similar motion patterns that belong to the same cluster.The proposed model can extract meaningful high-level features using only spatial-angular features obtained from refined tracklets(a.k.a‘trajectories’).We adopt l2-regularization and sparsity regularization,which can learn sparse representations of features,to guarantee the sparsity of the autoencoders.We employ the softmax layer to map the data points into accurate cluster representations.One of the best advantages of the SA-SSAE framework is it can manage VMS even when individuals move around randomly.This framework helps cluster the motion patterns effectively with higher accuracy.We put forward a new dataset with itsmanual ground truth,including 21 crowd videos.Experiments conducted on two crowd benchmarks demonstrate that the proposed model can more accurately group trajectories than the traditional clustering approaches used in previous studies.The proposed SA-SSAE framework achieved a 0.11 improvement in accuracy and a 0.13 improvement in the F-measure compared with the best current method using the CUHK dataset.

Action recognition using a hierarchy of feature groups

To improve the recognition performance of video human actions,an approach that models the video actions in a hierarchical way is proposed. This hierarchical model summarizes the action contents with different spatio-temporal domains according to the properties of human body movement.First,the temporal gradient combined with the constraint of coherent motion pattern is utilized to extract stable and dense motion features that are viewed as point features,then the mean-shift clustering algorithm with the adaptive scale kernel is used to label these features.After pooling the features with the same label to generate part-based representation,the visual word responses within one large scale volume are collected as video object representation.On the benchmark KTH（Kungliga Tekniska H？gskolan）and UCF （University of Central Florida）-sports action datasets,the experimental results show that the proposed method enhances the representative and discriminative power of action features, and improves recognition rates.Compared with other related literature,the proposed method obtains superior performance.

Anomaly detection in traffic surveillance with sparse topic model

Most research on anomaly detection has focused on event that is different from its spatial-temporal neighboring events.It is still a significant challenge to detect anomalies that involve multiple normal events interacting in an unusual pattern.In this work,a novel unsupervised method based on sparse topic model was proposed to capture motion patterns and detect anomalies in traffic surveillance.scale-invariant feature transform(SIFT)flow was used to improve the dense trajectory in order to extract interest points and the corresponding descriptors with less interference.For the purpose of strengthening the relationship of interest points on the same trajectory,the fisher kernel method was applied to obtain the representation of trajectory which was quantized into visual word.Then the sparse topic model was proposed to explore the latent motion patterns and achieve a sparse representation for the video scene.Finally,two anomaly detection algorithms were compared based on video clip detection and visual word analysis respectively.Experiments were conducted on QMUL Junction dataset and AVSS dataset.The results demonstrated the superior efficiency of the proposed method.

Review of biomechanical deviations among nonpregnant,pregnant,and postpartum cohorts

During pregnancy,women experience substantial changes in physiology,morphology,and hormonal systems.These changes have profound effects on the biomechanics of the human body,particularly the musculoskeletal system,resulting in discomfort,pain,and decreased body stability.Sufficient biomechanical knowledge is critical for understanding the etiology and precautions of musculoskeletal disorders.With awareness of health problems in the pregnant cohort,identification,intervention,and precaution of problems have garnered attention.Researchers have conducted studies to determine the biomechanics of pregnancy.There have been review studies on summarization.However,to the best of our knowledge,few studies have comprehensively described biomechanical changes throughout pre-,in-,and postpartum periods.This review analyzed available studies on biomechanical changes during these three periods in the electronic databases of PubMed,Scopus,and Cochrane from inception until June 2,2021.Synthesized the general information,age of the studied subjects,investigated periods,sample size,objectives,measurement tools,and outcomes of reviewed studies.And Using National Institutes of Health quality assessment tool for observational cohort and cross-sectional studies to assessment the quality of the reviewed articles.These studies revealed biomechanical deviations in body stability,motion patterns,and gait modes during these three periods.Regarding research content,there are insufficient studies on certain critical biomechanical aspects,such as the kinetic parameters of the inner body,which are the most direct factors related to musculoskeletal problems.According to the National Institutes of Health quality assessment tool for observational cohort and cross-sectional studies,a more comprehensive and explicit understanding of pregnancy biomechanics can be expected.

Design,analysis,and neural control of a bionic parallel mechanism

Although the torso plays an important role in the movement coordination and versatile locomotion of mammals,the structural design and neuromechanical control of a bionic torso have not been fully addressed.In this paper,a parallel mechanism is designed as a bionic torso to improve the agility,coordination,and diversity of robot locomotion.The mechanism consists of 6-degree of freedom actuated parallel joints and can perfectly simulate the bending and stretching of an animal’s torso during walking and running.The overall spatial motion performance of the parallel mechanism is improved by optimizing the structural parameters.Based on this structure,the rhythmic motion of the parallel mechanism is obtained by supporting state analysis.The neural control of the parallel mechanism is realized by constructing a neuromechanical network,which merges the rhythmic signals of the legs and generates the locomotion of the bionic parallel mechanism for different motion patterns.Experimental results show that the complete integrated system can be controlled in real time to achieve proper limb-torso coordination.This coordination enables several different motions with effectiveness and good performance.