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.

High-elevation Adaptation of Motion Visual Display Modifications in the Toad-Headed Agamid Lizards(Phrynocephalus)

Understanding the process of adaptation is a key mission in modern evolutionary biology.Animals living at high elevations face challenges in energy meta bolism due to several environmental constraints(e.g., oxygen supply, food availa bility,and movement time). Animal behavioral processes are intimately related to energy meta bolism, and therefore, behavioral modifica tions are expected to be an important mechanism for high-elevation adaptation. We tested this behavioral adaptation hypothesis using va ria tions of motion visual displays in toad-headed agamid lizards of the genus Phr ynocephalus. We predicted tha t complexity of visual motion displays would decrease with the increase of elevation, because motion visual displays are energetically costly. Displays of 12 Phr ynocephalus species were collected with elevations ranging from sea level to 4600 m. We quantified display complexity using the number of display components, display duration, pathways of display components, as well as display speed for each species. Association between display complexity and elevation was analyzed using the phylogenetic generalized least squares(PGLS)model. We found that both the number of display components and the average value of tail coil speed were negatively correlated with elevation, suggesting that toad-headed lizards living at high-elevation areas reduced their display complexity to cope with the environmental constraints. Our research provides direct evidence for high-elevation adaptation from a behavioral aspect and illustrates the potential impacts of environment heterogeneity on motion visual display diversification.

Multimodal collaborative BCI system based on the improved CSP feature extraction algorithm

Background As a novel approach for people to directly communicate with an external device,the study of brain-computer interfaces(BCIs)has become well-rounded.However,similar to the real-world scenario,where individuals are expected to work in groups,the BCI systems should be able to replicate group attributes.Methods We proposed a 4-order cumulants feature extraction method(CUM4-CSP)based on the common spatial patterns(CSP)algorithm.Simulation experiments conducted using motion visual evoked potentials(mVEP)EEG data verified the robustness of the proposed algorithm.In addition,to freely choose paradigms,we adopted the mVEP and steady-state visual evoked potential(SSVEP)paradigms and designed a multimodal collaborative BCI system based on the proposed CUM4-CSP algorithm.The feasibility of the proposed multimodal collaborative system framework was demonstrated using a multiplayer game controlling system that simultaneously facilitates the coordination and competitive control of two users on external devices.To verify the robustness of the proposed scheme,we recruited 30 subjects to conduct online game control experiments,and the results were statistically analyzed.Results The simulation results prove that the proposed CUM4-CSP algorithm has good noise immunity.The online experimental results indicate that the subjects could reliably perform the game confrontation operation with the selected BCI paradigm.Conclusions The proposed CUM4-CSP algorithm can effectively extract features from EEG data in a noisy environment.Additionally,the proposed scheme may provide a new solution for EEG-based group BCI research.

Form Properties of Moving Targets Bias Smooth Pursuit Target Selection in Monkeys

During natural viewing,we often recognize multiple objects,detect their motion,and select one object as the target to track.It remains to be determined how such behavior is guided by the integration of visual form and motion perception.To address this,we studied how monkeys made a choice to track moving targets with different forms by smooth pursuit eye movements in a two-target task.We found that pursuit responses were biased toward the motion direction of a target with a hole.By computing the relative weighting,we found that the target with a hole exhibited a larger weight for vector computation.The global hole feature dominated other form properties.This dominance failed to account for changes in pursuit responses to a target with different forms moving singly.These findings suggest that the integration of visual form and motion perception can reshape the competition in sensorimotor networks to guide behavioral selection.