Crowd Density Research Articles

Pose Mask: A Model-Based Augmentation Method for 2D Pose Estimation in Classroom Scenes Using Surveillance Images.

Solid developments have been seen in deep-learning-based pose estimation, but few works have explored performance in dense crowds, such as a classroom scene; furthermore, no specific knowledge is considered in the design of image augmentation for pose estimation. A masked autoencoder was shown to have a non-negligible capability in image reconstruction, where the masking mechanism that randomly drops patches forces the model to build unknown pixels from known pixels. Inspired by this self-supervised learning method, where the restoration of the feature loss induced by the mask is consistent with tackling the occlusion problem in classroom scenarios, we discovered that the transfer performance of the pre-trained weights could be used as a model-based augmentation to overcome the intractable occlusion in classroom pose estimation. In this study, we proposed a top-down pose estimation method that utilized the natural reconstruction capability of missing information of the MAE as an effective occluded image augmentation in a pose estimation task. The difference with the original MAE was that instead of using a 75% random mask ratio, we regarded the keypoint distribution probabilistic heatmap as a reference for masking, which we named Pose Mask. To test the performance of our method in heavily occluded classroom scenes, we collected a new dataset for pose estimation in classroom scenes named Class Pose and conducted many experiments, the results of which showed promising performance.

Estimating indoor crowd density and movement behavior using WiFi sensing

The fact that almost every person owns a smartphone device that can be precisely located is both empowering and worrying. If methods for accurate tracking of devices (and their owners) via WiFi probing are developed in a responsible way, they could be applied in many different fields, from data security to urban planning. Numerous approaches to data collection and analysis have been covered, some of which use active sensing equipment, while others rely on passive probing, which takes advantage of nearly universal smartphone usage and WiFi network coverage. In this study, we introduce a system that uses WiFi probing technologies aimed at tracking user locations and understanding individual behavior. We built our own devices to passively capture WiFi request probe packets from smartphones, without the phones being connected to the network. The devices were tested at the headquarters of the research sector of the Elm Company. The results of the analyses carried out to estimate the crowd density in offices and the flows of the crowd from one place to another are promising and illustrate the importance of such solutions in indoor and closed spaces.

The Feasibility of Integrating Wearable Cameras and Health Trackers for Measuring Personal Exposure to Urban Features

Built environment factors such as greenery, walkability, and crowd density are related to physical activity and mental health. New emerging wearable sensors provide an opportunity to objectively monitor human exposure to street-level urban features. However, very few studies have demonstrated how to objectively measure the association between the built environment, human emotions, and health. This pilot study proposes a new approach that employs a FrontRow wearable lifestyle camera, a GPS tracker, and an Empatica 4 wristband as a sensor package to track individuals during their everyday activities. Machine-learning methods are adopted to extract urban features. For this study, volunteers were asked to conduct a self-led city tour in Roskilde, Denmark, while using the wearable sensors. Study results demonstrate the feasibility of the proposed approach and the potential for using integrated, multi-sourced data in the study of urban health.

Crowd Density Estimation in Spatial and Temporal Distortion Environment Using Parallel Multi-Size Receptive Fields and Stack Ensemble Meta-Learning

The estimation of crowd density is crucial for applications such as autonomous driving, visual surveillance, crowd control, public space planning, and warning visually distracted drivers prior to an accident. Having strong translational, reflective, and scale symmetry, models for estimating the density of a crowd yield an encouraging result. However, dynamic scenes with perspective distortions and rapidly changing spatial and temporal domains still present obstacles. The main reasons for this are the dynamic nature of a scene and the difficulty of representing and incorporating the feature space of objects of varying sizes into a prediction model. To overcome the aforementioned issues, this paper proposes a parallel multi-size receptive field units framework that leverages the majority of the CNN layer’s features, allowing for the representation and participation in the model prediction of the features of objects of all sizes. The proposed method utilizes features generated from lower to higher layers. As a result, different object scales can be handled at different framework depths, and various environmental densities can be estimated. However, the inclusion of the vast majority of layer features in the prediction model has a number of negative effects on the prediction’s outcome. Asymmetric non-local attention and the channel weighting module of a feature map are proposed to handle noise and background details and re-weight each channel to make it more sensitive to important features while ignoring irrelevant ones, respectively. While the output predictions of some layers have high bias and low variance, those of other layers have low bias and high variance. Using stack ensemble meta-learning, we combine individual predictions made with lower-layer features and higher-layer features to improve prediction while balancing the tradeoff between bias and variance. The UCF CC 50 dataset and the ShanghaiTech dataset have both been subjected to extensive testing. The results of the experiments indicate that the proposed method is effective for dense distributions and objects of various sizes.

Anomalous event detection and localization in dense crowd scenes

Anomalous event detection and localization in dense crowd scenes

Exploring density rectification and domain adaption method for crowd counting.

Crowd counting has received increasing attention due to its important roles in multiple fields, such as social security, commercial applications, epidemic prevention and control. To this end, we explore two critical issues that seriously affect the performance of crowd counting including nonuniform crowd density distribution and cross-domain problems. Aiming at the nonuniform crowd density distribution issue, we propose a density rectifying network (DRNet) that consists of several dual-layer pyramid fusion modules (DPFM) and a density rectification map (DRmap) auxiliary learning module. The proposed DPFM is embedded into DRNet to integrate multi-scale crowd density features through dual-layer pyramid fusion. The devised DRmap auxiliary learning module further rectifies the incorrect crowd density estimation by adaptively weighting the initial crowd density maps. With respect to the cross-domain issue, we develop a domain adaptation method of randomly cutting mixed dual-domain images, which learns domain-invariance features and decreases the domain gap between the source domain and the target domain from global and local perspectives. Experimental results indicate that the devised DRNet achieves the best mean absolute error (MAE) and competitive mean squared error (MSE) compared with other excellent methods on four benchmark datasets. Additionally, a series of cross-domain experiments are conducted to demonstrate the effectiveness of the proposed domain adaption method. Significantly, when the A and B parts of the Shanghaitech dataset are the source domain and target domain respectively, the proposed domain adaption method decreases the MAE of DRNet by 47.6%.

Public Transport Crowding Valuation: Evidence from College Students in Guangzhou

Overcrowding has grown to be an increasingly important issue for public transit in Guangzhou, China. To capture traveler benefits of reduced crowding from improved public transport, it is necessary to identify the relevant importance of crowding to travelers. This paper analyzes the disutility of crowding in metro car and bus transit with discrete choice models. Based on the stated preference survey data from college students in Guangzhou, the results show that there is non-negligible impact of crowding on passenger travel. The relationship between the disutility of crowding and standee density is not linear; that is, the disutility increases at a modest rate as the standee density increases when it is easy to move around in cars and increases rapidly when it is difficult to move around in cars. Also, there is only a slight difference between the effects of crowding on metro and bus transit.

A systematic review for a highly accessible green roof for Malaysian public institution buildings

PurposeDespite its worldwide reputation as an effective solution to sustainable building development and energy efficiency, green roofs in Malaysian cities are rarely accessible. The architecture of the building primarily influences public accessibility, crime watch and safety level and events that encourage the public's engagement, which is evident in crowd density. The purpose of this paper is to discuss the social potential of highly accessible Malaysian green roofs as public space, initiated by the lack of local published material discussing on this topic.Design/methodology/approachThis study reviews the current issues concerning limited public accessibility on Malaysian public institution green roofs by systematic literature review and thematic analysis by comparing the effectiveness of applicable public space strategies on the green roof.FindingsThe criteria that have been identified and considered as study parameters include architecture, safety and surveillance, and active functions. Through systematic review of available literature, these characteristics contribute positively to public participation within the public realm.Social implicationsThe exploration of the social potential would establish a green roof as a thriving public space that welcomes the public from all ages and backgrounds, addressing the general public accessibility towards outdoor recreational areas, especially within dense urbanisation with diminishing green spaces.Originality/valueThis research highlights the key characteristics of the highly functional public space that could be applied in developing a guideline for designing future green roofs with high accessibility potential for the public in the city area, in parallel with the anticipated future growth in demand for green roofs infrastructure surrounding public buildings.

A Unified Multiple-Motion-Mode Framework for Socially Compliant Navigation in Dense Crowds

Mobile robots are expected to move safely and efficiently in socially-compliant ways in dense crowds. In this paper, a unified multiple-motion-mode framework is proposed to achieve socially-compliant navigation in dense crowds. The proposed framework consists of three successive phases, the identification of pedestrian groups and prediction of their trajectories, the generation of candidate local trajectories of the robot, and the determination of an optimal local trajectory. The pedestrians are grouped according to distances among pedestrians and their velocities and a predictor predicts the groups’ trajectories. Three robot motion modes, moving-solo, pedestrian-following, and courteous-stopping, are proposed based on a systematic analysis and classification of dense crowds and they are utilized to generate candidate local trajectories. A composite metric is proposed to determine the best local trajectory by simultaneously considering the robot motion safety, efficiency, and stability of the trajectory. The proposed robot navigation framework is evaluated both in simulations and real-world experiments with various scenarios. The results show that the proposed framework surpasses the state-of-the-art methods in terms of efficiency and social compliance. Note to Practitioners—This paper is motivated by the socially-compliant navigation problem of mobile robots in densely crowded environments such as hospitals and airports. Robots navigation in these scenes need to follow collective social conventions to ensure ordered and efficient public pedestrian traffics, and individual social conventions for acceptance by human society. Existing approaches generally consider only one type of social conventions cannot accomplish a truly socially-compliant navigation and cannot be naturally accepted in the human pedestrian environment. This paper proposes to imitate the pedestrian’s walking behavior for robot navigation by naturally choosing one of the three walking modes, moving-solo, following, and courteous-stopping. We construct the candidate trajectory sampling strategies for different motion modes based on the global path and the two types of social conventions. Experimental results show that the proposed framework can be applied in large-scale, densely crowded, unstructured, and human-robot coexisting environments for socially-compliant navigation. However, the proposed navigation framework has limited improvement in navigation efficiency in a highly crowded environment due to the limited feasible space. Even pedestrians are difficult to walk in these highly crowded environments. In future work, we will study how to auto-adjust the weights to steer the robot to efficient navigation in highly crowded environments.

Avoiding Dense and Dynamic Obstacles in Enclosed Spaces: Application to Moving in Crowds

This paper presents a closed-form approach to constrain a flow within a given volume and around objects. The flow is guaranteed to converge and to stop at a single fixed point. We show that the obstacle avoidance problem can be inverted to enforce that the flow remains enclosed within a volume defined by a polygonal surface. We formally guarantee that such a flow will never contact the boundaries of the enclosing volume and obstacles, and will asymptotically converge towards an attractor. We further create smooth motion fields around obstacles with edges (e.g. tables). Both obstacles and enclosures may be time-varying, i.e. moving, expanding and shrinking. The technique enables a robot to navigate within an enclosed corridor while avoiding static and moving obstacles. It was applied on an autonomous robot (QOLO) in a static complex indoor environment, and also tested in simulations with dense crowds. The final proof of concept was performed in an outdoor environment in Lausanne. The QOLO-robot successfully traversed a marketplace in the center of town in presence of a diverse crowd with a non-uniform motion pattern.

Dynamic Crowd Accident-Risk Assessment Based on Internal Energy and Information Entropy for Large-Scale Crowd Flow Considering COVID-19 Epidemic

With the increase in inevitable large-scale crowd aggregation, disastrous pedestrian stampedes occurred with increasing frequency over the past decade. To prevent these tragedies, it is significant to assess crowd accident-risk (CAR) and identify high-risk areas to control crowd flow dynamically. The cost function of a conventional fluid dynamics model is improved with new items of Gaussian white noise and protection factor, considering both the abnormal pedestrian movements and social distance control due to epidemic, thereby to establish an improved crowd flow model comprehensively. Different from conventional density-based pedestrian aggregation-risk models, this study proposes a hybrid crowd accident-risk assessment (HCRA) model based on internal energy and information entropy. Using the HCRA model, we can consider not only crowd density but also the modulus and direction of a crowd velocity vector simultaneously. Then this study designs a framework to realize crowd accident risk assessment based on the improved crowd-flow model and HCRA model. To validate the proposed models, case studies of CAR assessment in the large-scale waiting hall of the Shanghai Hongqiao railway station are conducted. The pedestrian social control distance-range of 1.0 m–2.0 m under the COVID-19 epidemic situation is verified numerically. Moreover, a valuable result is that this social control distance-range can be shortened to 1.0 m–1.9 m without increase of crow accident-risk. Subsequently, the down-limit of accommodation-capacity of this large waiting hall can be enhanced to 10.54% under this epidemic.

NFOMP: Neural Field for Optimal Motion Planner of Differential Drive Robots With Nonholonomic Constraints

Optimal motion planning is one of the most critical problems in mobile robotics. On the one hand, classical sampling-based methods propose asymptotically optimal solutions to this problem. However, these planners cannot achieve smooth and short trajectories in reasonable calculation time. On the other hand, optimization-based methods are able to generate smooth and plain trajectories in a variety of scenarios, including a dense human crowd. However, modern optimization-based methods use the precomputed signed distance function for collision loss estimation, and it limits the application of these methods for general configuration spaces, including a differential drive non-circular robot with non-holonomic constraints. Moreover, optimization-based methods lack the ability to handle U-shaped or thin obstacles accurately. We propose to improve the optimization methods in two aspects. Firstly, we developed an obstacle neural field model to estimate collision loss; training this model together with trajectory optimization allows improving collision loss continuously, while achieving more feasible and smoother trajectories. Secondly, we forced the trajectory to consider non-holonomic constraints by adding Lagrange multipliers to the trajectory loss function. We applied our method for solving the optimal motion planning problem for differential drive robots with non-holonomic constraints, benchmarked our solution, and proved that the novel planner generates smooth, short, and plain trajectories perfectly suitable for a robot to follow, and outperforms the state-of-the-art approaches by 25% on normalized curvature and by 75% on the number of cusps in the MovingAI environment.

Cross-Domain Attention Network for Unsupervised Domain Adaptation Crowd Counting

Unsupervised domain adaptation crowd counting (UDACC) has been studied with practical research utility by getting rid of the labeling burden on large-scale dense crowds in the target domain. Current methods generalize well within the specific domain gap by directly aligning domain distributions or translating synthetic data to realistic images. However, it is difficult to define domain gaps among complex real-world datasets, in which the images vary greatly in style, density level and/or content. To tackle this problem, in this paper, we propose a Cross-Domain Attention Network (CDANet), which can effectively generalize the model to the unlabeled domain on both unsupervised synthetic-to-realistic and realistic-to-realistic crowd counting. Specifically, we propose a Cross-Domain Attention Module (CDAM) to learn domain-related information between the source and target domain, which extracts relations in cross-domain attentive information, thus enhancing crowd-informative features. Moreover, to make our CDAM invariant to domain shifts, we introduce a consistency penalty to ensure that the attention maps are consistent before and after the domain shifting. Thus our CDANet can pay attention to the shared counting information across domains, while remaining its invariant ability during domain adaptation. Extensive experiments on several common benchmarks for UDACC demonstrate that our CDANet gets competitive results on both unsupervised synthetic-to-realistic and realistic-to-realistic UDACC tasks.

Survey on algorithms of people counting in dense crowd and crowd density estimation

Survey on algorithms of people counting in dense crowd and crowd density estimation

Social rather than physical crowding reduces the required interpersonal distance in virtual environments.

Interpersonal distance plays an important role in human social interaction. With the increasing usage of virtual reality in social interaction, people's interpersonal distance in virtual space attracts great attention. It remains unclear whether and to what extent human-required interpersonal distance is altered by crowded virtual scenes. In this study, we manipulated crowd density in virtual environments and used the classical stop-distance paradigm to measure required interpersonal distances at different crowd densities. We found that people's required interpersonal distance decreased with increased social crowdedness but not with physical crowdedness. Moreover, the decrease of two types of interpersonal distance was associated with the globally averaged crowd density rather than local crowd density. The reduction is not due to the imitation of other virtual humans in the crowd. Moreover, we developed a model to describe the quantitative relationships between the crowdedness of the environment and the required interpersonal distance. Our finding provides insights into designing user-friendly virtual humans in metaverse virtual worlds.

Research on 24-Hour Dense Crowd Counting and Object Detection System Based on Multimodal Image Optimization Feature Fusion

Motivation. In the environment of day and night video surveillance, in order to improve the accuracy of machine vision dense crowd counting and target detection, this paper designs a day and night dual-purpose crowd counting and crowd detection network based on multimode image fusion. Methods. Two sub-models, RGBD-Net and RGBT-Net, are designed in this paper. The depth image features and thermal imaging features are effectively fused with the features of visible light images, so that the model has stronger anti-interference characteristics and robustness to the light noise interference caused by the sudden fall of light at night. The above models use density map regression-guided detection method to complete population counting and detection. Results. The model completed daytime training and testing on MICC dataset. Through verification, the average absolute error of the model was 1.025, the mean square error was 1.521, and the recall rate of target detection was 97.11%. Night vision training and testing were completed on the RGBT-CC dataset. After verification, the average absolute error of the network was 18.16, the mean square error was 32.14, and the recall rate of target detection was 97.65%. By verifying the effectiveness of the multimode medium-term fusion network, it is found to exceed the current most advanced bimodal fusion method. Conclusion. The experimental results show that the proposed multimodal fusion network can solve the counting and detection problem in the video surveillance environment during day and night. The ablation experiment further proves the effectiveness of the parameters of the two models.

Deep Learning driven automated person detection and tracking model on surveillance videos

Recently, person detection and tracking in a video scene of a surveillance system were grabbing higher interest because of its extensive range of applications in gender classification, abnormal event recognition, person identification, human gait characterization, individual counting in a dense crowd, and fall detection for older people, etc. Several methodologies were implied for person detection for surveillance applications. This paper presents a new DL driven automated person detection and tracking (DLD-APDT) model on surveillance videos. The major goal of the proposed DLD-APDT model is to identify people and track them in the videos. To accomplish this, the proposed DLD-APDT model initially performs frame conversion process where the input video is converted into a set of frames. In addition, the proposed DLD-APDT model employs EfficientDet model as object detector, i.e., persons and track them. EfficientDet is a recently developed object detector that makes use of different optimization and backbone tweaks, like bi-directional feature pyramid network (BiFPN) and a compound scaling approach. Moreover, root means square propagation (RMSProp) optimizer is applied to optimally choose the hyperparameters related to the EfficientNet model which helps in accomplishing enhanced performance. The experimental validation of the DLD-APDT model takes place using two training datasets namely PascalVOC and PenFudan datasets. The simulation results reported the enhanced detection and tracking performance of the DLD-APDT technique over recent approaches.

Spatial–Temporal Convolutional Model for Urban Crowd Density Prediction Based on Mobile-Phone Signaling Data

Urban crowd density prediction is essential for transport demand management and public safety monitoring. Existing studies for crowd density prediction only focus on a few transport modes which do not cover all urban crowds. Moreover, most existing models rely on grid-based divisions, which cannot accommodate unevenly distributed populations. This study proposes a novel model for predicting crowd density in urban areas with mobile-phone signaling data (MPSD). MPSD contain diversified travel and activity information that is not limited to specific transport modes and are practical for activity-travel research. We introduce methods for preprocessing MPSD and extracting information on crowd density and activity choice behaviors (such as activity location, activity type, and activity duration). A novel spatial-temporal convolutional model is proposed for urban crowd density prediction. We design an attention-based feature fusion block, enabling the model to comprehensively consider historical crowd density, activity choices, and external factors (such as weather conditions and holidays). With tailored spatial-temporal convolution blocks, our proposed model successfully captures spatial-temporal information related to crowd movement in different scenes. Unlike existing grid-based models, the proposed model can predict crowd density in irregular-shaped regions. Experimental results show that our model outperforms baseline models that do not apply activity information when selecting different prediction offsets, especially for long-term predictions. The correlation between prediction error and the regularity of crowd activities is also analyzed. The subsequent parameter analysis shows that the influence of external factors on urban crowd movement varies according to activity choices and other travel-related factors like commuting distance.

A modified cellular automaton model of pedestrian evacuation in a tunnel fire

This paper presents evacuation simulations of occupants’ flow in a tunnel fire via a modified cellular automaton model and puts forward corresponding optimized evacuation strategies. A 2-D grid field represents the simulated tunnel and a 20 MW fire is placed in the middle of the tunnel. Fire, as the key factor is regarded both as a dynamic obstacle along with its development and a repelling force on each occupant. During occupants’ movement, game theory is taken into consideration when people intend to move to the same target simultaneously and they could either corporate or fight for the target. Seven exits layouts and the moving conflict preference of pedestrians during movement are investigated with three levels of crowd densities namely low, medium and high in conditions either with fire or without fire. The results show frequent conflict among occupants and single, overlaid exits will result in a longer evacuation and a declined evacuation efficiency. The decline is very sensitive to crowds’ density and the high crowd density performs a much worse evacuation efficiency. In addition, the fire aggravates the decline to a large extent in all conditions and it would lead to a larger possibility of injury for occupants during evacuation in a tunnel fire. Therefore, it is necessary to predict evacuation flow prudently with models considering fire rather than those without fire, which may result in an unrealistic optimistic result.

Revealing the determinants of the intermodal transfer ratio between metro and bus systems considering spatial variations

Buses and metros are two main public transit modes, and these modes are crucial components of sustainable transportation systems. Promoting reciprocal integration between bus and metro systems requires a deep understanding of the effects of multiple factors on transfers among integrated public transportation transfer modes, i.e., metro-to-bus and bus-to-metro. This study aims to reveal the determinants of the transfer ratio between bus and metro systems and quantify the associated impacts. The transfer ratio between buses and metros is identified based on large-scale transaction data from automated fare collection systems. Meanwhile, various influencing factors, including weather, socioeconomic, the intensity of business activities, and built environment factors, are obtained from multivariate sources. A multivariate regression model is used to investigate the associations between the transfer ratio and multiple factors. The results show that the transfer ratio of the two modes significantly increases under high temperature, strong wind, rainfall, and low visibility. The morning peak hours attract a transfer ratio of up to 57.95%, and the average hourly transfer volume is 0.94 to 1.38 times higher at this time than in other periods. The intensity of business activities has the most significant impact on the transfer ratio, which is approximately 1.5 to 15 times that of the other independent variables. Moreover, an adaptative geographically weighted regression is utilized to investigate the spatial divergences of the influences of critical factors on the transfer ratio. The results indicate that the impact of a factor presents spatial heterogeneity and even shows opposite effects (in terms of positive and negative) on the transfer ratio in different urban contexts. For example, among the related socioeconomic variables, the impact of the housing price on the downtown transfer ratio is larger than that in the suburbs. Crowd density positively influences the transfer ratio at most stations in the northern region, whereas it shows negative results in the southern region. These findings provide valuable insights for public transportation management and promote the effective integration of bus and metro systems to provide enhanced transfer services.