arXiv subjects

Armin Seyfried

Publications and source records attributed to Armin Seyfried.

At least 19 recordsLinked to original sources

arXiv2026

A well-motivated model of pedestrian dynamics

In pedestrian dynamics, the internal drive that propels individuals toward their goals is typically captured by a single, fixed parameter, the desired walking speed. This simplification overlooks that motivation fluctuates in response to changing spatial and social conditions within a crowd. This paper proposes a dynamic motivation model grounded in expectancy-value theory from psychology, in which each agent's motivation evolves over time depending on proximity to the goal, relative position among other pedestrians, and individual goal importance. The resulting motivation modulates multiple movement parameters simultaneously, including walking speed, gap-closing behavior, and interpersonal spacing. The model is evaluated in simulated pre-bottleneck waiting scenarios using paired statistical comparisons across multiple random seeds and population sizes, and compared with trajectory data from the CROMA concert-entry bottleneck experiments under low- and high-motivation framings. Simulations show that the dynamic model produces structured heterogeneity in the crowd: agents self-organize into differentiated positions near the bottleneck, with those closer to the front occupying less space, a pattern absent in the static baseline but clearly present in the experimental data. These findings suggest that motivation in crowds should be understood not as a uniform increase in urgency, but as a mechanism that reorganizes competitive positioning along spatial and social axes. Future work should extend the framework to open-door throughput scenarios, larger populations, and richer social interactions such as group cohesion and cooperative strategies.

physics.soc-ph↗

arXiv2026

When legs and bodies synchronize: Two-level collective dynamics in dense crowds

Ultra-dense crowds, in which physical contact between people cannot be avoided, pose major safety concerns. Nevertheless, the underlying dynamics driving their collective behaviours remain poorly understood. Existing dense crowd models, mostly two-dimensional and contact-based, overlook biomechanical mechanisms that govern individual balance motion. In this study, we introduce a minimal two-level pedestrian model that couples upper body and legs dynamics, allowing us to capture transitions between balanced and unbalanced states at the individual scale. Whereas previous models fail to achieve it, this coupling gives rise to emergent collective behaviours observed empirically, such as self-organized waves and large-scale rotational motion within the crowd. The model bridges basic individual biomechanical concepts and macroscopic flow dynamics, offering a new framework for modelling and understanding collective motions in ultra-dense crowds.

physics.soc-ph↗

arXiv2025

Modelling vehicle and pedestrian collective dynamics: Challenges and advances

In our urbanised societies, the management and regulation of traffic and pedestrian flows is of considerable interest for public safety, economic development, and the conservation of the environment. However, modelling and controlling the collective dynamics of vehicles and pedestrians raises several challenges. Not only are the individual entities self-propelled and hard to describe, but their complex nonlinear physical and social interactions makes the multi-agent problem of crowd and traffic flow even more involved. In this chapter, we purport to review the suitability and limitations of classical modelling approaches through four examples of collective behaviour: stop-and-go waves in traffic flow, lane formation, long-term avoidance behaviour, and load balancing in pedestrian dynamics. While stop-and-go dynamics and lane formation can both be addressed by basic reactive models (at least to some extent), the latter two require anticipation and/or coordination at the level of the group. The results highlight the limitations of classical force-based models, but also the need for long-term anticipation mechanisms and multiscale modelling approaches. In response, we review new developments and modelling concepts.

physics.soc-ph↗

arXiv2025

Social LSTM with Dynamic Occupancy Modeling for Realistic Pedestrian Trajectory Prediction

In dynamic and crowded environments, realistic pedestrian trajectory prediction remains a challenging task due to the complex nature of human motion and the mutual influences among individuals. Deep learning models have recently achieved promising results by implicitly learning such patterns from 2D trajectory data. However, most approaches treat pedestrians as point entities, ignoring the physical space that each person occupies. To address these limitations, this paper proposes a novel deep learning model that enhances the Social LSTM with a new Dynamic Occupied Space loss function. This loss function guides Social LSTM in learning to avoid realistic collisions without increasing displacement error across different crowd densities, ranging from low to high, in both homogeneous and heterogeneous density settings. Such a function achieves this by combining the average displacement error with a new collision penalty that is sensitive to scene density and individual spatial occupancy. For efficient training and evaluation, five datasets were generated from real pedestrian trajectories recorded during the Festival of Lights in Lyon 2022. Four datasets represent homogeneous crowd conditions -- low, medium, high, and very high density -- while the fifth corresponds to a heterogeneous density distribution. The experimental findings indicate that the proposed model not only lowers collision rates but also enhances displacement prediction accuracy in each dataset. Specifically, the model achieves up to a 31% reduction in the collision rate and reduces the average displacement error and the final displacement error by 5% and 6%, respectively, on average across all datasets compared to the baseline. Moreover, the proposed model consistently outperforms several state-of-the-art deep learning models across most test sets.

cs.CV↗

arXiv2025

Pedestrian Flow Analysis in High-Density Crowds: Continuity Equation with Voronoi-Based Fields

Since the beginning of the century, capturing trajectories of pedestrian streams precisely from video recordings has been possible. To enable measurements at high density, the heads of the pedestrians are marked and tracked, thus providing a complete representation of the phase space. However, classical definitions of flow, density, and velocity of pedestrian streams are based on different segments in phase space. In addition, traditional methods fail with high densities of people, as heads move even when a crowd is blocked and standing still. In this article, Voronoi decomposition is used to construct density and velocity fields from pedestrian trajectories to solve this problem. Combined with the continuity equation, a flow equation on the basis of trajectories is derived satisfying the conservation of particle numbers exactly. The proposed method allows definitions of all quantities in the same segment of phase space even on scales smaller than the dimensions of a pedestrian. It is shown that these new definitions of flow, density, velocity are consistent with classical measurements and make it possible to determine standstill in pedestrian flows even when individual body parts are moving. These properties allow to scrutinize inconsistencies in the state of the art of pedestrian fundamental diagrams.

physics.soc-ph↗

arXiv2025

Learning Extremely High Density Crowds as Active Matters

Video-based high-density crowd analysis and prediction has been a long-standing topic in computer vision. It is notoriously difficult due to, but not limited to, the lack of high-quality data and complex crowd dynamics. Consequently, it has been relatively under studied. In this paper, we propose a new approach that aims to learn from in-the-wild videos, often with low quality where it is difficult to track individuals or count heads. The key novelty is a new physics prior to model crowd dynamics. We model high-density crowds as active matter, a continumm with active particles subject to stochastic forces, named 'crowd material'. Our physics model is combined with neural networks, resulting in a neural stochastic differential equation system which can mimic the complex crowd dynamics. Due to the lack of similar research, we adapt a range of existing methods which are close to ours for comparison. Through exhaustive evaluation, we show our model outperforms existing methods in analyzing and forecasting extremely high-density crowds. Furthermore, since our model is a continuous-time physics model, it can be used for simulation and analysis, providing strong interpretability. This is categorically different from most deep learning methods, which are discrete-time models and black-boxes.

cs.CV↗

arXiv2024

Human Motion Prediction under Unexpected Perturbation

We investigate a new task in human motion prediction, which is predicting motions under unexpected physical perturbation potentially involving multiple people. Compared with existing research, this task involves predicting less controlled, unpremeditated and pure reactive motions in response to external impact and how such motions can propagate through people. It brings new challenges such as data scarcity and predicting complex interactions. To this end, we propose a new method capitalizing differential physics and deep neural networks, leading to an explicit Latent Differential Physics (LDP) model. Through experiments, we demonstrate that LDP has high data efficiency, outstanding prediction accuracy, strong generalizability and good explainability. Since there is no similar research, a comprehensive comparison with 11 adapted baselines from several relevant domains is conducted, showing LDP outperforming existing research both quantitatively and qualitatively, improving prediction accuracy by as much as 70%, and demonstrating significantly stronger generalization.

cs.CV↗

arXiv2024

Temporal segmentation of motion propagation in response to an external impulse

In high-density crowds, local motion can propagate, amplify, and lead to macroscopic phenomena, including 'density waves'. These density waves only occur when individuals interact, and impulses are transferred to neighbours. How this impulse is passed on by the human body and which effects this has on individuals is still not fully understood. To further investigate this, experiments focusing on the propagation of a push were conducted. In the experiments the crowd is greatly simplified by five people lining up in a row. The rearmost person in the row was pushed forward in a controlled manner with a punching bag. The intensity of the push, the initial distance between participants and the initial arm posture were varied. Collected data included side view and top view video recordings, head trajectories, 3D motion using motion capturing (MoCap) suits as well as pressure measured at the punching bag. With a hybrid tracking algorithm, the MoCap data are combined with the head trajectories to allow an analysis of the motion of each limb in relation to other persons. The observed motion of the body in response to the push can be divided into three phases. These are (i) receiving an impulse, (ii) receiving and passing on an impulse, and (iii) passing on an impulse. Using the 3D MoCap data, we can identify the start and end times of each phase. To determine when a push is passed on, the forward motion of the person in front has to be considered. The projection of the center of mass relative to the initial position of the feet is a measure of the extent to which a person is displaced from the rest position. Specifying the timing of these phases is particularly important to distinguish between different types of physical interactions. Our results contribute to the development and validation of a pedestrian model for identifying risks due to motion propagation in dense crowds.

physics.soc-ph↗

arXiv2023

A Novel Voronoi-based Convolutional Neural Network Framework for Pushing Person Detection in Crowd Videos

Analyzing the microscopic dynamics of pushing behavior within crowds can offer valuable insights into crowd patterns and interactions. By identifying instances of pushing in crowd videos, a deeper understanding of when, where, and why such behavior occurs can be achieved. This knowledge is crucial to creating more effective crowd management strategies, optimizing crowd flow, and enhancing overall crowd experiences. However, manually identifying pushing behavior at the microscopic level is challenging, and the existing automatic approaches cannot detect such microscopic behavior. Thus, this article introduces a novel automatic framework for identifying pushing in videos of crowds on a microscopic level. The framework comprises two main components: i) Feature extraction and ii) Video labeling. In the feature extraction component, a new Voronoi-based method is developed for determining the local regions associated with each person in the input video. Subsequently, these regions are fed into EfficientNetV1B0 Convolutional Neural Network to extract the deep features of each person over time. In the second component, a combination of a fully connected layer with a Sigmoid activation function is employed to analyze these deep features and annotate the individuals involved in pushing within the video. The framework is trained and evaluated on a new dataset created using six real-world experiments, including their corresponding ground truths. The experimental findings indicate that the suggested framework outperforms seven baseline methods that are employed for comparative analysis purposes.

cs.CV↗

arXiv2023

Inside a Life-Threatening Crowd: Analysis of the Love Parade Disaster from the Perspective of Eyewitnesses

During the Love Parade disaster in 2011 in Duisburg, Germany, twenty one visitors were killed and more than five hundred injured in a very dense crowd on the route to and from the festival area. Approximately nine hundred visitors who had been among this crowd were subsequently interviewed by police officers as eyewitnesses. This paper content analyses a random sample of 136 of these witness statements, focusing on how those present perceived the crowd, how they behaved, how they experienced the event emotionally, what happened to their bodies, and which collective dynamics they described. This approach provides a perspective from within crowd dynamics which are usually observed from a top-view perspective. Almost all the attendees became strongly focused on the staircase and the pole the only visible ways out of the crowd. In some cases, they tried to reach these destinations by all means and at the expense of others. But at the same time, helping behavior is the behavior most frequently mentioned. Although witnesses described feelings of intense fear, they reject the idea of mass panic. As the most dangerous dynamics, a combination of falls (often after people had fainted) and transversal waves (which made people fall on top of each other) could be reconstructed. When people fall in a tightly packed crowd, a hole can form which pulls in more people due to the pressure on those standing around the edge of the hole, thus creating a pile of wedged bodies.

physics.soc-ph↗

arXiv2023

A Cloud-based Deep Learning Framework for Early Detection of Pushing at Crowded Event Entrances

Crowding at the entrances of large events may lead to critical and life-threatening situations, particularly when people start pushing each other to reach the event faster. Automatic and timely identification of pushing behavior would help organizers and security forces to intervene early and mitigate dangerous situations. In this paper, we propose a cloud-based deep learning framework for automatic early detection of pushing in crowded event entrances. The proposed framework initially modifies and trains the EfficientNetV2B0 Convolutional Neural Network model. Subsequently, it integrates the adapted model with an accurate and fast pre-trained deep optical flow model with the color wheel method to analyze video streams and identify pushing patches in real-time. Moreover, the framework uses live capturing technology and a cloud-based environment to collect video streams of crowds in real-time and provide early-stage results. A novel dataset is generated based on five real-world experiments and their associated ground truth data to train the adapted EfficientNetV2B0 model. The experimental setups simulated a crowded event entrance, while the ground truths for each video experiment was generated manually by social psychologists. Several experiments on the videos and the generated dataset are carried out to evaluate the accuracy and annotation delay time of the proposed framework. The experimental results show that the proposed framework identified pushing behaviors with an accuracy rate of 87% within a reasonable delay time.

cs.LG↗

arXiv2023

Identification of social groups and waiting pedestrians at railway platforms using trajectory data

To investigate the impact of social groups on waiting behaviour of passengers at railway platforms a method to identify social groups through the monitoring of distances between pedestrians and the stability of those distances over time is introduced. The method allows the recognition of groups using trajectories only and thus opens up the possibility of studying crowds in public places without constrains caused by privacy protection issues. Trajectories from a railway platform in Switzerland were used to analyse the waiting behaviour of passengers in dependence of waiting time as well as the size of social groups. The analysis of the trajectories shows that the portion of passengers travelling in groups reaches up to 10\% during the week and increases to 20 \% on the weekends. 60\% of the groups were pairs, larger groups were less frequent. With increasing group size, the mean speed of the members decreases. Individuals and pairs often choose waiting spots at the sides of the stairs and in vicinity of obstacles, while larger groups wait close to the platform entries. The results indicate that passengers choose waiting places according to the following criteria and ranking: shortest ways, direction of the next intended action, undisturbed places and ensured communication. While individual passengers often wait in places where they are undisturbed and do not hinder others, the dominating comfort criterion for groups is to ensure communication. The results regarding space requirements of waiting passengers could be used for different applications. E.g. to enhance the level of service concept assessing the comfort of different types of users, to avoid temporary bottlenecks to improve the boarding and alighting process or to increase the robustness of the performance of railway platforms during peak loads by optimising the pedestrian distribution.

physics.soc-ph↗

arXiv2023

Pedestrian Crowd Management Experiments: A Data Guidance Paper

Understanding pedestrian dynamics and the interaction of pedestrians with their environment is crucial to the safe and comfortable design of pedestrian facilities. Experiments offer the opportunity to explore the influence of individual factors. In the context of the project CroMa (Crowd Management in transport infrastructures), experiments were conducted with about 1000 participants to test various physical and social psychological hypotheses focusing on people's behaviour at railway stations and crowd management measures. The following experiments were performed: i) Train Platform Experiment, ii) Crowd Management Experiment, iii) Single-File Experiment, iv) Personal Space Experiment, v) Boarding and Alighting Experiment, vi) Bottleneck Experiment and vii) Tiny Box Experiment. This paper describes the basic planning and implementation steps, outlines all experiments with parameters, geometries, applied sensor technologies and pre- and post-processing steps. All data can be found in the pedestrian dynamics data archive.

physics.soc-ph↗

arXiv2023

Influence of Gender Composition in Pedestrian Single-File Experiments

Various studies address the question of what factors are relevant to the course of the fundamental diagram in single-file experiments. Some indicate that there are differences due to group composition when gender is taken into account. For this reason, further single-file experiments with homogeneous and heterogeneous group compositions were conducted. A Tukey HSD test was performed to investigate whether there are differences between the mean of velocity in different density ranges. A comparison of different group compositions shows that the effect of gender can only be seen, if at all, in a small density interval. Regression analyses were also conducted to determine whether, at high densities, the distance between individuals depends on the gender of the neighboring pedestrians and to establish what human factors have an effect on the velocity. An analysis of the distances between individuals at high densities indicates that there is no effect of the gender of the neighboring pedestrians. Taking into account additional human factors in a regression analysis does not improve the model.

physics.soc-ph↗

arXiv2022

Influence of individual factors on fundamental diagrams of pedestrians

In recent years, numerous studies have been published dealing with the effect of individual characteristics of pedestrians on the fundamental diagram. These studies compared cumulative data on individuals in a group homogeneous in terms of one human factor such as age but heterogeneous in terms of other factors for instance gender. In order to examine the effect of all determined as well as undetermined human factors, individual fundamental diagrams are introduced and analyzed using multiple linear regression. A single-file school experiment with students of different age, gender, and height is therefore considered. Single individuals appearing in different runs are analyzed to study the effect of human factors such as height, age and gender and all other unknown individual effects such as motivation or attention to the individual speed. The analysis shows that for students age and height are strongly correlated and, consequently, age can be ignored. Furthermore, the study shows that gender has a weak effect and other nonmeasurable individual characteristics have a stronger effect than height. In a further step, a mixed model is used as well as the multiple linear model. Here, it is shown that the mixed model that considers all other unknown individual effects of each person as a random factor is preferable to the model where the individual speed only depends on the variables of headway, height, and all other unknown individual effects as fixed factors.

physics.soc-ph↗

arXiv2021

Anticipation in a velocity-based model for pedestrian dynamics

Lane formation in bidirectional pedestrian streams is based on a stimulus-response mechanism and strategies of navigation in a fast-changing environment. Although microscopic models that only guarantee volume exclusion can qualitatively reproduce this phenomenon, they are not sufficient for a quantitative description. To quantitatively describe this phenomenon, a minimal anticipatory collision-free velocity model is introduced. Compared to the original velocity model, the new model reduces the occurrence of gridlocks and reproduces the movement of pedestrians more realistically. For a quantitative description of the phenomenon, the definition of an order parameter is used to describe the formation of lanes at transient states and to show that the proposed model compares relatively well with experimental data. Furthermore, the model is validated by the experimental fundamental diagrams of bidirectional flows.

physics.soc-ph↗

arXiv2021

Prolonged Clogs in Bottleneck Simulations for Pedestrian Dynamics

This article studies clogging phenomena using a velocity-based model for pedestrian dynamics. First, a method to identify prolonged clogs in simulations was introduced. Then bottleneck simulations were implemented with different initial and boundary conditions. The number of prolonged clogs was analyzed to investigate the decisive factors causing this phenomenon. Moreover, the time lapse between two consecutive agents passing the exit, and the trajectories of agents were analyzed. The influence of three types of factors was studied: parameters of the spatial boundaries, algorithmic factors related to the implementation of the model, and the movement model. Parameters of the spatial boundaries include the width and position of the bottleneck exit. Algorithmic factors are the update methods and the size of the time step. Model parameters cover several parameters describing the level of motivation, the strength and range of impact among agents, and the shape of agents. The results show that the occurrence of prolonged clogs is closely linked to parameters of the spatial boundaries and the movement model but has virtually no correlation with algorithmic factors.

physics.soc-ph↗

arXiv2020

Crowds in front of bottlenecks at entrances from the perspective of physics and social psychology

This article presents an interdisciplinary study of physical and social psychological effects on crowd dynamics based on a series of bottleneck experiments. Bottlenecks are of particular interest for applications such as crowd management and design of emergency routes because they limit the performance of a facility. In addition to previous work on the dynamics within the bottleneck, this study focuses on the dynamics in front of the bottleneck, more specifically, at entrances. The experimental setup simulates an entrance scenario to a concert consisting of an entrance gate (serving as bottleneck) and a corridor formed by barriers. The parameters examined are the corridor width, degree of motivation and priming of the social norm of queuing. The analysis is based on head trajectories and questionnaires. We show that the density of persons per square metre depends on motivation and also increases continuously with increasing corridor width meaning that a density reduction can be achieved by a reduction of space. In comparison to other corridor widths observed, the narrowest corridor is rated as being fairer, more comfortable and as showing less unfair behaviour. Pushing behaviour is seen as ambivalent: it is rated as unfair and listed as a strategy for faster access.

physics.soc-ph↗