Search Results (3,579)

Pedestrian trajectory prediction is vital for crowd analysis and human–-robot interaction. Recent deep models enhance accuracy by modeling social interactions and scene context, but they often remain opaque and rarely address privacy risks associated with learning individualized motion patterns. We propose CAMP, a Context-Aware, Multimodal, and Privacy-preserving pedestrian trajectory prediction framework designed around a role-aligned multimodal architecture, in which trajectory representations, dynamic scene cues, and explicit spatial interaction constraints are modeled through complementary branches. In CAMP, the trajectory encoder separates shared motion regularities from individualized motion tendencies, the optical-flow encoder captures motion-centric transient scene dynamics, and the potential-field encoder provides an interpretable spatial cost prior for obstacle avoidance and social interaction modeling. A Transformer-based decoder fuses these modalities to predict future trajectory distributions. To reduce the exposure of personalized motion patterns, we apply targeted DP-SGD only to the individual branch during the private fine-tuning stage, while treating the remaining frozen components as post-processing under the stated threat model. Experiments on the ETH/UCY benchmark show that CAMP achieves competitive ADE/FDE performance under the reported setting, while its private variant DP-CAMP maintains a reasonable utility–privacy trade-off across several reported privacy budgets. Full article

The large-scale generation of asphalt dust waste (ADW) has raised increasing environmental concerns, while its high-value utilization in cementitious materials remains insufficiently explored, particularly in terms of mechanical performance, durability-related properties, and integrated sustainability evaluation. In this study, a graded utilization strategy based on particle size was proposed to incorporate ADW into alkali-activated concrete paving blocks, in which fine ADW fraction (<0.075 mm) was used as a partial replacement of blast furnace slag (BFS), while the coarser ADW fraction was used as a partial replacement of river sand, aiming at sustainable pavement applications. In addition, two types of ADW with different lithologies, namely limestone ADW and basalt ADW, along with their combined system, were investigated. The results show that the incorporation of ADW effectively enhances the engineering performance of paving blocks. The compressive strength increased from 45.3 MPa to 56.6 MPa, while water absorption decreased from 5.3% to 4.1%. All mixtures satisfied the requirements for abrasion resistance and slip resistance, demonstrating their compliance with the performance criteria for pedestrian pavement applications. Among all mixtures, the combined use of limestone ADW and basalt ADW exhibited the best overall performance. The improved performance may be attributed to the combined effects of graded particle utilization and the potential compositional complementarity between calcium-rich limestone ADW and silica–alumina-rich basalt ADW, which is consistent with the denser microstructure observed in SEM images. In addition, the proposed strategy contributes to improved solid waste utilization and reduced consumption of natural resources, as reflected in the quantitative sustainability assessment. Overall, this study demonstrates that graded utilization of ADW is a feasible approach for developing alkali-activated paving materials, with promising performance and sustainability potential. Full article

Most established street-vitality assessment tools were developed in temperate, predominantly Western urban settings and therefore do not adequately capture the climatic and socio-spatial conditions of hot–arid cities. This study develops and validates the Resident-Centered Street Vitality Framework (RCSVF) using Riyadh as a case study representative of the Arabian Desert urban context. Drawing on a cross-sectional quantitative design, the research integrates a resident survey across nineteen neighborhoods (N = 1102), physical observations of 133 street segments, a visual preference survey (N = 418), and GIS-based spatial analysis. The results reveal marked intra-urban inequality in perceived environmental quality and demonstrate that service proximity is a substantially stronger predictor of residential satisfaction than street physical quality alone. Residents consistently rated shading, green space, and pedestrian infrastructure as the weakest dimensions of their neighborhoods. These findings indicate that street vitality in hot–arid settings cannot be validly assessed through imported observer-based metrics. A resident-centered, climate-responsive framework is required to capture how thermal exposure, functional accessibility, and everyday social use interact in shaping street experience. Full article

The use of plastic waste as a filler in concrete, particularly in paving block production, represents an approach aligned with circular economy principles. While previous studies have focused on mechanical properties, the effect of such materials on slip risk remains insufficiently investigated, especially for pedestrian applications. This study evaluates the influence of the volumetric content of recycled plastic waste from electrical cable insulation on slip resistance of concrete paving blocks. A series of specimens was prepared with 0–45% replacement of natural aggregate by granulated cable insulation (GCI). Slip resistance was measured using the British Pendulum Tester and expressed as Skid Resistance Value (SRV) after statistical processing. Two sliders were used, Mounted Shoe 55 and Mounted Shoe 96, corresponding to road and pedestrian conditions. The results show that increasing GCI content reduces mass by approximately 9.6 g per 1% GCI, reaching a reduction of about 20% at 50% GCI. For polished surfaces, SRV increased up to 77 (MS55) and 75 (MS96) at 40–45% GCI. For ground surfaces, optimal performance was observed at 10% GCI, while higher contents reduced SRV and caused mechanical degradation above 30–35% GCI. The results indicate that optimized GCI content can improve slip resistance while reducing material weight. Full article

Street soundscapes significantly shape communities’ environmental perceptions, behaviour and urban sustainability. Previous research has mainly focused on physical and acoustic aspects, while limited attention has been given to emotional and behavioural dimensions. This study explores how expert participants perceive street soundscapes through personal, physical, behavioural, and emotional dimensions, using international online focus groups with soundscape experts, urban planners, and policymakers (n = 12). Analysis followed a deductive thematic approach establishing four main a priori themes, with additional inductive coding used to refine these themes. The findings reveal that perception is shaped by contextual, cultural, temporal, multisensory, and environmental affordance factors. Notably, silence was found to carry a dilemma—perceived as either safe or unsafe depending on pedestrian density—and religious and cultural soundmarks were identified as evoking place attachment and belonging, areas largely overlooked in existing literature. These soundscapes were associated with emotional responses, including comfort, safety, restoration, and belonging, and with pedestrian behaviour encompassing mobility choices, coping strategies, and social interactions. Furthermore, seven out of ten Healthy Streets metrics were directly referenced by participants, highlighting the close relationship between acoustic environments and healthy streets design. Future studies should examine cultural, temporal, and spatial street characteristics and their effects on human behaviour and emotional responses. Full article

High-resolution GPS trajectories pose a geospatial processing challenge: transforming temporally ordered observations into structured spatial representations that retain intra-trip state transitions at metropolitan scale. This study develops and validates a scalable geospatial transformation workflow for detecting and structuring recurrent mid-trip stops from large-scale trajectory data. Using approximately 97 million GPS observations from Seoul’s public bike-sharing system, stopping episodes are identified through speed-based segmentation and density-based spatial clustering (DBSCAN). Recurrent stopping hotspots are attributed with spatial context via a land-use overlay and proximity analysis to pedestrian crossings. Sequential transitions between recurrent hotspots are represented as directed and weighted hotspot-to-hotspot networks, whose structural properties are evaluated using connectivity, clustering, path length, and modularity metrics under degree-preserving randomization. The workflow emphasizes explicit parameterization and modular processing, aligning with reproducible GIS-based spatial analytical frameworks. By converting fine-grained trajectory observations into validated mesoscopic connectivity representations, the framework provides a transferable geospatial processing pipeline for extracting structured connectivity information from high-resolution trajectory datasets. Full article

With the continuous expansion of building scale and increasingly complex functions, pedestrian congestion within buildings has become increasingly prominent. To identify high-utilization spaces during the architectural design stage, so as to optimize the design at the source and effectively alleviate the risk of oversaturated spatial flow in super high-rise buildings, transportation hubs and other complex buildings, this paper investigates spatial flow estimation methods in the architectural design phase. First, based on detailed data from microscopic pedestrian simulation in buildings, this study conducts an in-depth analysis of the factors influencing spatial flow from the perspectives of spatial structure and pedestrian demand. Then, by incorporating the visual integration degree from space syntax and the proposed Origin-Destination (OD) influence range definition method that accounts for obstacle detours, a spatial utilization intensity index integrating both factors is developed. Furthermore, regression analysis is employed to achieve spatial flow estimation based on the constructed spatial utilization intensity index. Finally, taking the basement 1 floor of the Shenzhen Bay Super Headquarters Base C Tower connected with the metro as an example, the effectiveness of the proposed method is verified. The results show that the MAPE is 26%. This study provides an effective estimation method for the prediction of spatial flow in the architectural design stage. Full article

Background/Objectives: A floating hip injury, defined as an ipsilateral fracture of the pelvis or acetabulum combined with a femoral fracture, represents a rare and devastating musculoskeletal injury resulting from high-energy trauma. Although Müller type A floating hip injuries comprising an acetabular fracture with an ipsilateral femoral fracture are recognized for their clinical complexity, the long-term prognostic factors influencing functional outcomes remain poorly elucidated. This study aimed to identify independent prognostic factors associated with unsatisfactory long-term functional outcomes in patients with Müller type A floating hip injuries. Methods: A retrospective study was performed on 68 consecutive patients with Müller type A floating hip injuries who underwent surgical fixation at a single tertiary trauma center, with a minimum follow-up period of 5 years. Functional outcomes were assessed using the Majeed score, and patients were dichotomized into satisfactory (n = 48; 70.6%) and unsatisfactory (n = 20; 29.4%) outcome groups. Acetabular fractures were classified according to the Judet–Letournel system, and femoral fractures were classified by fracture level (proximal, shaft, or distal). Radiological outcomes were evaluated using Matta’s radiological grading system. Demographic, injury-specific, and treatment-related variables were compared between groups using the Mann–Whitney U test and chi-square test with Bonferroni correction. A multivariate binary logistic regression model was constructed to determine independent predictors of unsatisfactory outcomes. Results: The mean age was 37.15 ± 12.07 years, with a male predominance (67.6%). The predominant mechanism of injury was pedestrian struck by vehicle (54.4%), followed by motor vehicle collision (27.9%) and fall from height (17.6%); collectively, high-energy vehicular trauma accounted for 82.3% of cases. In the univariate analysis, transverse with posterior wall acetabular fracture pattern (p = 0.001), proximal femur fracture level (p = 0.001), associated lower extremity fractures (p = 0.001), nerve damage (p = 0.001), higher body mass index (BMI) (p = 0.001), and lower Matta’s radiological scores (p = 0.001) were significantly associated with unsatisfactory outcomes. Three independent predictors emerged in the multivariate logistic regression: BMI (OR = 1.50; 95% CI: 1.05–2.15; p = 0.025), the presence of associated lower extremity fractures (OR = 29.02; 95% CI: 2.83–297.67; p = 0.005), and Matta’s radiological score (OR = 0.06; 95% CI: 0.01–0.56; p = 0.014). The model yielded internal discriminatory metrics within the acceptable range (overall accuracy 89.7%, sensitivity 95.8%, specificity 75.0%, Nagelkerke R² = 0.757); however, given the limited events-per-variable ratio (~6.7) and the wide confidence intervals observed for some predictors, these internal performance estimates are likely optimistic due to potential overfitting, and the findings should be interpreted as exploratory pending external validation. Conclusions: Elevated BMI, the presence of associated ipsilateral lower extremity fractures, and poor quality of acetabular reduction, assessed via Matta’s radiological criteria, are independent determinants of unsatisfactory long-term functional outcomes in Müller type A floating hip injuries. These findings underscore the critical importance of achieving anatomical reduction in the acetabulum and highlight the compounding effect of additional ipsilateral limb injuries on patient prognosis. Full article

The development of metro-led urban underground public spaces (UUPSs) provides urban residents with extensive pedestrian-friendly activity areas sheltered from rain, snow, strong winds, and other extreme weather conditions. Although an increasing number of people are engaging in daily commercial and leisure activities within UUPSs, problems such as inconvenient transfer, poor visibility, and a lack of natural light, which indicate poor environmental quality, have led to an uneven distribution of user behavior, thereby reducing the efficiency of space utilization. Our aim in this study was to predict UUPS utilization rates by investigating the relationship between UUPS environmental attributes and user behavior characteristics and preferences. Six typical UUPSs in Wuhan were selected as case studies. User behavior data were collected using panoramic camera recordings, on-site observations, and space syntax methods, while spatial environmental factors were quantified. The correlation between various factors and multi-dimensional user behavior characteristics was discussed, and a Random Forest model was established to predict behavioral preferences. Our results indicate that accessibility and visibility are fundamental factors influencing user behavior characteristics, while the impact of landscape elements is relatively low. Regarding behavioral preference prediction, UUPS environmental features achieved the highest prediction accuracy for leisure behaviors, whereas the predictive performance for sports activities was lower. In this study, we reveal the influence of UUPS environmental factors on user behavior characteristics and predict preference patterns of different behaviors for space types. Focusing on the behavioral needs of space users, we provide a reference for the subsequent human-centered design of UUPSs. Full article

Reliable pedestrian perception in autonomous driving depends not only on detecting the target, but also on how completely and consistently its three-dimensional geometry is captured from different sensor viewpoints. This study presents a LiDAR-based observability framework for evaluating pedestrian representation on the ANTA compact autonomous vehicle platform using a roof-mounted Top LiDAR (TL), a Front-Right LiDAR (FRL), and their fused configuration. The pedestrian was analyzed in a canonical local frame using geometric extent ratios, projected surface occupancy, voxel-based volumetric occupancy, and statistical descriptors of the local point distribution, integrated into a global observability score, $S_{3D}$. A Distance-Robustness Index (DRI), an overlap-based complementarity analysis, and a lightweight temporal centroid-sensitivity check over 20 consecutive frames were used to characterize performance across distance. Using ROS 2 bag data processed offline in MATLAB R2025b the fused configuration achieved the highest mean global score (0.563), compared with 0.504 for FRL and 0.432 for TL, and the highest robustness ($\mathrm{DRI}=0.5628$, $CV=10.7\%$). The results show that 1 m maximizes local density, 2–3 m maximize projected and volumetric completeness, and 7 m provides the best balanced observability. Within the evaluated platform and under the controlled benchmark conditions, complementary multi-LiDAR fusion provided the strongest overall geometry-aware pedestrian representation. Full article

Many place recognition approaches, which identify previously visited places or locations by matching current sensory data, such as 2D RGB images and 3D point clouds, have been proposed to achieve accurate and robust localization and loop closure detection in global positioning system (GPS)-denied environments. Since visual place recognition (VPR) methods that rely on images captured by camera sensors are highly sensitive to variations in appearance, including changes in lighting, surface color, and shadows, they can lead to poor place recognition accuracy. In contrast, light detection and ranging (LiDAR)-based place recognition (LPR) approaches based on 3D point cloud data that captures the shape and geometric structure of the environment are robust to changes in place appearance and can therefore provide more reliable place recognition results than VPR methods. This work presents an indoor LPR method called PointNetVLAD-based indoor pedestrian localization (PIPL). PIPL is a deep network model that uses PointNetVLAD to learn to extract global descriptors from 3D LiDAR point cloud data. PIPL can recognize places previously visited by a pedestrian using point clouds captured by a low-cost LiDAR sensor on a smartphone in small-scale indoor environments, while PointNetVLAD performs place recognition for vehicles using high-cost LiDAR, GPS, and inertial measurement unit (IMU) sensors in large-scale outdoor areas. For place recognition on 3D point cloud reference maps generated from LiDAR scans, PointNetVLAD exploits the universal transverse mercator (UTM) coordinate system based on GPS and IMU measurements, whereas PIPL uses a virtual coordinate system designed in this study due to the unavailability of GPS indoors. In experiments conducted in campus buildings, PIPL shows significant advantages over NetVLAD (known as a convolutional neural network (CNN)-based VPR method). Particularly in indoor environments with repetitive scenes where geometric structures are preserved and image-based appearance features are sparse or unclear, PIPL achieved $39\%$ higher top-1 accuracy and $10\%$ higher top-3 accuracy compared to NetVLAD. Furthermore, PIPL achieved place recognition accuracy comparable to NetVLAD even with a small number of points in a 3D point cloud and outperformed NetVLAD even with a smaller model training dataset. The experimental results also indicate that PIPL requires over $76\%$ less place retrieval time than NetVLAD while maintaining robust place classification performance. Full article

With the rapid advancement of autonomous driving in modern society, dense pedestrian detection technology has encountered performance bottlenecks. To address this, we propose a robust and lightweight pedestrian detection algorithm, RMP-YOLO, designed to efficiently detect small, occluded, and low-light objects. Firstly, RFAConv is utilized as the core component of the backbone network, combining standard convolution with attention mechanisms and using group convolution to extract features from the spatial receptive field. Secondly, MobileViTv3 is introduced into the backbone to combine CNNs with Transformers. The model is further enhanced by adjusting feature fusion, introducing residual connections, and optimizing local representation with deep convolutional layers. Finally, the PIoUv2 loss function is employed for bounding-box regression, significantly reducing detection errors for small-scale pedestrians in crowded environments. Experimental results demonstrate that RMP-YOLO improves mAP@0.5 by 1.3% on a custom dataset and 0.91% on the WiderPerson dataset. Crucially, it maintains high efficiency with only 3.71 million parameters and 6.29 GFLOPs, meeting the deployment requirements for low computational power and high precision. Full article

The Camino de Santiago, a UNESCO-listed pilgrimage route, has experienced sustained growth in visitor numbers, challenging municipalities to preserve cultural integrity while ensuring service quality. This study reviews people-counting technologies and proposes a smart pilgrim management framework grounded in flux measurement systems to support data-driven and sustainable decision-making. Drawing on the smart tourism literature, the conceptual framework integrates infrared counters, mobile tracking solutions, and GPS/Wi-Fi data to generate real-time insights into pilgrim flows. A pilot simulation illustrates how these data can inform operational and strategic planning. The framework enables local authorities to monitor pedestrian movements, anticipate service demands (sanitation, accommodation, and safety), and detect overcrowding in sensitive heritage areas. By incorporating technological solutions into traditionally low-tech pilgrimage settings, municipalities can transition from reactive to proactive management approaches. The paper contributes a scalable and ethically grounded framework tailored to heritage pilgrimage routes, advancing smart tourism applications in culturally significant contexts. Full article

Pulse-width-modulated (PWM) automotive headlights enhance nighttime event-based camera detection, yet systematic parameter optimization for vulnerable road user detection remains unexplored. This study evaluates PWM frequency, duty cycle, light distribution, ego-vehicle speed, and ambient lighting under European New Car Assessment Programme-inspired crossing scenarios for cyclist and pedestrian detection. Results establish performance ranging from substantial improvements to severe degradation relative to continuous illumination. Cyclist detection achieves robust performance with high-frequency modulation across light distributions, while low-frequency operation with low beam produces severe degradation through background noise accumulation. Pedestrian detection requires high beam with street lighting enabled; low beam universally fails regardless of modulation parameters. Limited parameter combinations achieve simultaneous improvements for both targets. Detection performs optimally on retroreflective surfaces, while low-reflectivity clothing limits capability, requiring target-specific optimization. Full article

Existing studies of street-related emotional perception mainly rely on static scene evaluations, which cannot capture the cumulative effects of environmental exposure during continuous walking. To address this limitation, this study proposes a method for predicting pedestrian physiological responses in sequential audiovisual street environments. Four real-world walking routes were selected, with outbound and return directions treated as independent paths, yielding eight paths and 32 valid samples. EEG, ECG, sound pressure level, first-person video, and GPS data were synchronously collected to construct a 1 s multimodal time-series dataset. Pearson correlation, Kendall correlation, and mutual information analyses were used to examine linear, monotonic, and nonlinear relationships between environmental variables and physiological indicators, and the resulting weights were incorporated into a Long Short-Term Memory (LSTM) model for multi-step prediction. Visual elements and noise exposure were the main factors influencing physiological responses. Among the models, the mutual-information-weighted LSTM performed best, achieving an R² of 0.77 for heart rate variability (RMSSD), whereas prediction of the EEG ratio (β/α and θ/β) remained limited. An additional independent street sample outside the training set was then used to generate a dual-dimensional EEG-ECG physiological response map, demonstrating the model’s potential for identifying emotional risk segments and supporting street-level micro-renewal. Full article