The DOI matches its Crossref record; 23 of the 25 references it cites were found in a registry or an index.
You'll never walk alone: Modeling social behavior for multi-target tracking
Abstract
Object tracking typically relies on a dynamic model to predict the object's location from its past trajectory. In crowded scenarios a strong dynamic model is particularly important, because more accurate predictions allow for smaller search regions, which greatly simplifies data association. Traditional dynamic models predict the location for each target solely based on its own history, without taking into account the remaining scene objects. Collisions are resolved only when they happen. Such an approach ignores important aspects of human behavior: people are driven by their future destination, take into account their environment, anticipate collisions, and adjust their trajectories at an early stage in order to avoid them. In this work, we introduce a model of dynamic social behavior, inspired by models developed for crowd simulation. The model is trained with videos recorded from birds-eye view at busy locations, and applied as a motion model for multi-people tracking from a vehicle-mounted camera. Experiments on real sequences show that accounting for social interactions and scene knowledge improves tracking performance, especially during occlusions.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Histograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584
Social force model for pedestrian dynamics.
Helbing, D., & Molnár, P. (1995) · Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics · doi:10.1103/physreve.51.4282 · Cited by 7,299
A discriminatively trained, multiscale, deformable part model
Felzenszwalb, P. F., McAllester, D., & Ramanan, D. (2008) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2008.4587597 · Cited by 2,972
Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics
Bernardin, K., & Stiefelhagen, R. (2008) · EURASIP Journal on Image and Video Processing · doi:10.1155/2008/246309 · Cited by 2,918
Abnormal crowd behavior detection using social force model
Mehran, R., Oyama, A., & Shah, M. A. (2009) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2009.5206641 · Cited by 1,698
Lerner, A., Chrysanthou, Y., & Lischinski, D. (2007) · Computer Graphics Forum · doi:10.1111/j.1467-8659.2007.01089.x · Cited by 1,199
A Boosted Particle Filter: Multitarget Detection and Tracking
Okuma, K., Taleghani, A., de Freitas, N., Little, J. J., & Lowe, D. G. (2004) · Lecture notes in computer science · doi:10.1007/978-3-540-24670-1_3 · Cited by 1,025
Gräbner, H., & Bischof, H. (2006) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2006.215 · Cited by 998
Global data association for multi-object tracking using network flows
Zhang, L., Li, Y., & Nevatia, R. (2008) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2008.4587584 · Cited by 945
People-tracking-by-detection and people-detection-by-tracking
Andriluka, M., Roth, S., & Schiele, B. (2008) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2008.4587583 · Cited by 825
Flagged first, then by citations
2 of the 25 references couldn’t be matched to a known paper.
Research works printed but not linked: 1
- schadschneider, 2001, Cellular automaton approach to pedestrian dynamics—theory, PED
Cited by
Sorted by citations
Items 701–800 of 1,000
Improving Pedestrian Prediction Models With Self-Supervised Continual Learning
Knoedler, L., Salmi, C., Zhu, H., et al. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3148475 · Cited by 14
“Albeit being efficient, these models are usually trained and evaluated using (offline) bench-marking datasets [20], [21], [22], [23], which limits their online adaption to unseen scenarios.”
MethodPredicting and recognizing human interactions in public spaces
Poiesi, F., & Cavallaro, A. (2014) · Journal of Real-Time Image Processing · doi:10.1007/s11554-014-0428-8 · Cited by 14
Robust Trajectory Forecasting for Multiple Intelligent Agents in Dynamic ScenePreprint
Zhu, Y., Ren, D., Fan, M., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.13133 · Cited by 14
“Because there is no HD map information and the planned trajectory for the ego-agent, there is no ego-trajectory feature, ego-motion feature and environment feature in the proposed method for the ETH & UCY, and the ApolloScape datasets.”
BackgroundInfluentialMirzabagheri, A., Ahmadi, M. A., Zhang, N., Alirezaee, R., Mozaffari, S., & Alirezaee, S. (2025) · Vehicles · doi:10.3390/vehicles7020057 · Cited by 14
Dong, Y., Wang, L., Zhou, S., et al. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3594116 · Cited by 13
“ETH [57] and UCY [58] are the most widely used benchmark datasets for pedestrian trajectory prediction.”
High-density crowd behaviors segmentation based on dynamical systems
Zhang, D., Xu, J., Sun, M., et al. (2016) · Multimedia Systems · doi:10.1007/s00530-016-0520-y · Cited by 13
“For example, Johnson [11] uses tracking methods for individuals, Marques [12] uses them for groups of pedestrians, and the more recent work of Pellegrini [13] uses tracking based on a social force model.”
MethodHuman trajectory forecasting using a flow-based generative model
Zhang, B., Wang, T., Zhou, C., et al. (2022) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.105236 · Cited by 13
Wu, W., Yi, W., Wang, X., & Zheng, X. (2023) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2023.104154 · Cited by 13
MetaTraj: Meta-Learning for Cross-Scene Cross-Object Trajectory Prediction
Shi, X., Zhang, H., Yuan, W., et al. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3299112 · Cited by 13
“To evaluate MetaTraj’s performance when transferring to other scenes and objects, we test it on various prediction models over three publicly available datasets: UCY [30], ETH [31] and another large dataset ApolloScape [32].”
MethodAlghodhaifi, H., & Lakshmanan, S. (2023) · Sensors · doi:10.3390/s23177361 · Cited by 13
“Moreover, we trained our model on additional datasets, including ETH, UCY, HOTEL, ZARA1, and ZARA2.”
MethodInfluentialModality Exploration, Retrieval and Adaptation for Trajectory Prediction
Sun, J., Li, Y., Chai, L., & Lu, C. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3316389 · Cited by 13
“Weconduct exhaustive experiments on four widely-used benchmarks, ETH [20] &UCY [21], Stanford Drone Dataset [22], NBA Sports VU Dataset [23], and Argoverse [24], covering different application scenarios.”
MethodPredicting pedestrian trajectories at different densities: A multi-criteria empirical analysis
Korbmacher, R., Dang, H. S., & Tordeux, A. (2023) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2023.129440 · Cited by 13
“The substantial error of the CV approach on the ETH dataset highlights its challenging nature, with the highest deviation from keeping speed and direction constant.”
BackgroundMethodInfluentialLearning Adaptive Neighborhoods for Graph Neural Networks
Saha, A., Méndez, O., Russell, C., et al. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.02060 · Cited by 13
“ETH [27] and UCY [18] — 5 subsets of widely used real-world pedestrian trajectories.”
BackgroundLearning Nonlinear Dynamics of Ocean Surface Vessel With Multistep Constraints
Wang, T., Skulstad, R., Kanazawa, M., et al. (2024) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2024.3396359 · Cited by 13
“Regarding trajectory prediction, the model performance is usually assessed with the average displacement error (ADE), and the final displacement error (FDE) proposed by [28].”
MethodSocialTrans: Transformer based social intentions interaction for pedestrian trajectory prediction
Chen, K., Zhao, X., Huang, Y., et al. (2025) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2025.130435 · Cited by 13
Event Based Switched Dynamic Bayesian Networks for Autonomous Cognitive Crowd Monitoring
Chiappino, S., Marcenaro, L., Morerio, P., et al. (2012) · Augmented vision and reality · doi:10.1007/8612_2012_8 · Cited by 13
“More recently [26], [27], [28], [29], a social force model describing the interactions among the individual members of a group of people has been proposed to detect abnormal events in crowd videos.”
BackgroundCrowd Event Perception Based on Spatio-temporal Viscous Fluid Field
Su, H., Yang, H., Zheng, S., et al. (2012) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2012.32 · Cited by 13
“The representative works include the methods proposed by Jacques [3] and Pellegrini [8], in which the Voronoi diagram and linear trajectory avoidance are utilized to recognize the crowd behavior, respectively.”
MethodAction Sequence Predictions of Vehicles in Urban Environments using Map and Social Context
Zaech, J.-N., Dai, D.-X., Liniger, A., & van Gool, L. (2020) · IEEE/RJS International Conference on Intelligent RObots and Systems · doi:10.1109/iros45743.2020.9340643 · Cited by 16
“[17] show that social interactions and scene knowledge can boost tracking performance.”
BackgroundAG-GAN: An Attentive Group-Aware GAN for pedestrian trajectory prediction
Song, Y., Bisagno, N., Hassan, S. Z., & Conci, N. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9413077 · Cited by 13
“We test our approach on real-world videos from two publicly available datasets, commonly used as benchmarks, namely the UCY dataset [14] and the ETH dataset [34].”
MethodEgocentric Two-frame Pedestrian Trajectory Prediction Algorithm Based on a Panoramic Camera
Long, J., Mei, J., & Ma, G. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3225018 · Cited by 14
“The datasets used for pedestrian trajectory prediction, such as ETH [5] and UCY [6], are mostly top-down views, which are not consistent with the application scenario of this article.”
MethodChen, K., Zhu, H., Tang, D., et al. (2023) · Image and Vision Computing · doi:10.1016/j.imavis.2023.104671 · Cited by 13
Generic Object Crowd Tracking by Multi-Task Learning
Luo, W., & Kim, T. (2013) · British Machine Vision Conference (BMVC) · doi:10.5244/c.27.73 · Cited by 13
“Methods belonging to the former category derive robust appearance models [18], delicate motion models and interaction models [32] [42] and develop a cost function considering multiple types of information up to the current frame and estimate the lowest cost state [6, 14, 24, 3 36, 46].”
BackgroundUncertainty Models for TTC-Based Collision-Avoidance
Forootaninia, Z., Karamouzas, I., & Narain, R. (2017) · doi:10.15607/rss.2017.xiii.002 · Cited by 13
“The uncertainty could also be estimated on the fly using methods such as LTA [23] and BRVO [21] which estimate the motion of other agents from noisy data, using a pre-trained pedestrian model or ensemble Kalman filtering respectively.”
MethodRETRACTED ARTICLE: Multipoint infrared laser-based detection and tracking for people counting
Wu, H., Gao, C., Cui, Y., et al. (2017) · Neural Computing and Applications · doi:10.1007/s00521-017-3196-0 · Cited by 13
“R A In order to solve these problems, some works tend to investigate human motion analysis, scene level and behavioral factors, which affect spatial arrangements and movements of people [4, 17, 31].”
RetractedBisagno, N., Conci, N., & Zhang, B. (2017) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2017.8078494 · Cited by 13
Human trajectory prediction for automatic guided vehicle with recurrent neural network
Song, C., Chen, Z., Qi, X., Zhao, B., Hu, Y., Liu, S., & Zhang, J. (2018) · The Journal of Engineering · doi:10.1049/joe.2018.8264 · Cited by 14
RoadTrack: Realtime Tracking of Road Agents in Dense and Heterogeneous Environments
Chandra, R., Bhattacharya, U., Randhavane, T., Bera, A., & Manocha, D. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9196612 · Cited by 13
Multimodal multi-pedestrian path prediction for autonomous cars
Poibrenski, A., Klusch, M., Vozniak, I., et al. (2021) · ACM SIGAPP Applied Computing Review · doi:10.1145/3447332.3447333 · Cited by 13
“For example, in [48] the authors propose a linear trajectory avoidance model, and in [66] the social force model is utilized.”
Destination Flow for Crowd Simulation
Pellegrini, S., Gall, J., Sigal, L., et al. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33885-4_17 · Cited by 13
“To simulate crowds we use the steering model presented in [14], termed LTA.”
BackgroundMethodInfluentialSelf-citationGAMMA: A General Agent Motion Prediction Model for Autonomous DrivingPreprint
Luo, Y., & Cai, P. (2019) · arXiv (Cornell University) · Cited by 13
CausalCity: Complex Simulations with Agency for Causal Discovery and ReasoningPreprint
McDuff, D., Song, Y., Lee, J., Vineet, V., Vemprala, S. H., Gydé, N., Salman, H., Ma, S., Sohn, K., & Kapoor, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.13364 · Cited by 13
“In such cases, the interaction between agents can be modelled using social forces, velocity obstacles [45], or linear trajectory avoidance [46].”
Social robot navigation: a review and benchmarking of learning-based methods
Alyassi, R., Cadena, C., Riener, R., & GRANADOS, D. F. P. (2025) · Frontiers in Robotics and AI · doi:10.3389/frobt.2025.1658643 · Cited by 14
A pedestrian trajectory prediction method based on improved LSTM network
Liu, F., Duan, S., & Juan, W. (2023) · IET Image Processing · doi:10.1049/ipr2.12954 · Cited by 12
Hafeezallah, A., Al-Dhamari, A., & Abu–Bakar, S. A. R. (2024) · International Journal of Disaster Risk Reduction · doi:10.1016/j.ijdrr.2024.104262 · Cited by 12
Knowledge-aware Graph Transformer for Pedestrian Trajectory Prediction
Liu, Y., Zhang, Y., Li, K., et al. (2023) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc57777.2023.10421989 · Cited by 12
“In this work, the proposed model is trained and tested on two public pedestrian trajectory datasets: ETH [27] and UCY [28].”
MethodYang, X., Wang, S., Zhu, Y., Zhou, D., & Li, T. (2024) · Information Sciences · doi:10.1016/j.ins.2024.120433 · Cited by 12
SocialCVAE: Predicting Pedestrian Trajectory via Interaction Conditioned Latents
Xiang, W., Yin, H., Wang, H., & Jin, X. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i6.28439 · Cited by 15
A Unified Environmental Network for Pedestrian Trajectory Prediction
Su, Y., Li, Y., Wang, W. W., et al. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i5.28301 · Cited by 12
“While in the long-term prediction, our method is totally superior to Y-net on ETH, because multimodal motion behaviors in the long-term prediction are obvious and the advantage of modeling diverse motions at non-goal time steps can be presented.”
MethodInfluentialModeling interpretable social interactions for pedestrian trajectory
Liu, Q., Shi, X., Jiang, R., et al. (2024) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2024.104617 · Cited by 12
Integrating deep learning model and virtual reality technology for motion prediction in emergencies
Li, M., Fanfan, P., Zhang, Y., Chen, T., & Du, H. (2024) · Safety Science · doi:10.1016/j.ssci.2024.106721 · Cited by 12
Wang, P., Pan, M., Liu, Z., et al. (2024) · Journal of Marine Science and Engineering · doi:10.3390/jmse12122233 · Cited by 12
Mizuta, K., & Leung, K. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802549 · Cited by 25
“We trained CoBL-Diffusion using the ETH pedestrian dataset [46] for 500 epochs on 276,874 8-second trajectories.”
PROBABILISTIC MULTI-PERSON TRACKING USING DYNAMIC BAYES NETWORKS
Klinger, T., Rottensteiner, F., & Heipke, C. (2015) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprsannals-ii-3-w5-435-2015 · Cited by 12
On the Reliability of LSTM-MDL Models for Pedestrian Trajectory Prediction
Hug, R., Becker, S., Hübner, W., & Arens, M. (2019) · Communications in computer and information science · doi:10.1007/978-3-030-19816-9_2 · Cited by 14
“Dataset UCY[17] UCY EWAP[16] EWAP SSD[18] Details Students Zara ETH Hotel Hyang”
BackgroundMethodInfluentialSILA: An Incremental Learning Approach for Pedestrian Trajectory Prediction
Habibi, G., Jaipuria, N., & How, J. P. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00520 · Cited by 12
“The datasets used for this set of experiments are: ETH [16] and UCY [30].”
BackgroundMethodLearning Occupancy Priors of Human Motion From Semantic Maps of Urban Environments
Rudenko, A., Palmieri, L., Doellinger, J., Lilienthal, A. J., & Arras, K. O. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3062010 · Cited by 12
“Examples include person and group tracking [15]–[17], in par-”
BackgroundEnhancing trajectory prediction using sparse outputs: application to team sports
Victor, B., Nibali, A., He, Z., et al. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-05888-w · Cited by 12
“Two common datasets used for pedestrian trajectory prediction are ETH [27] and UCY [22].”
MethodRoadside pedestrian motion prediction using Bayesian methods and particle filter
Xu, Q., Wu, H., Wang, J., et al. (2021) · IET Intelligent Transport Systems · doi:10.1049/itr2.12090 · Cited by 12
Multi-Camera Trajectory Forecasting With Trajectory Tensors.
Styles, O., Guha, T., & Sánchez, V. (2022) · PubMed · doi:10.1109/tpami.2021.3107958 · Cited by 12
STENet: A hybrid spatio-temporal embedding network for human trajectory forecasting
Zhang, B., Yuan, C., Wang, T., et al. (2021) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2021.104487 · Cited by 12
Zhu, Y., Ren, D., Xu, Y., et al. (2021) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3466182 · Cited by 12
“The ETH [30] and UCY [23] are benchmarks for pedestrian trajectory prediction.”
BackgroundExploring Attention GAN for Vehicle Motion Prediction
Gómez-Huélamo, C., Conde, M. V., Ortiz, M., Montiel-Marín, S., Barea, R., & Bergasa, L. M. (2022) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc55140.2022.9921804 · Cited by 13
“Prior knowledge on motion prediction in pedestrian datasets [11] [12] usually focuses on Deep Learning methods such as Long Short-Term Memory (LSTM) [8] and Generative Adversarial Network (GAN) [13].”
Self-supervised Social Relation Representation for Human Group Detection
Li, J., Han, R., Yan, H., et al. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19833-5_9 · Cited by 12
Human Context: Modeling Human-Human Interactions for Monocular 3D Pose Estimation
Andriluka, M., & Sigal, L. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-31567-1_26 · Cited by 12
“Recent tracking-by-detection (TbD) methods [1] have shown impressive results in real world scenarios, but with a few exceptions [2], are restricted to individual people performing simple cyclic activities (e.”
BackgroundAnimating agents based on radial view in crowd simulation
Li, F., & Wong, S. (2016) · doi:10.1145/2993369.2993373 · Cited by 12
Identity Retention of Multiple Objects under Extreme Occlusion Scenarios using Feature Descriptors
Israni, D., & Mewada, H. K. (2018) · Journal of Communications Software and Systems · doi:10.24138/jcomss.v14i4.541 · Cited by 12
Human Intention Recognition for Human Aware Planning in Integrated Warehouse Systems
Petković, T., Hvězda, J., Rybecký, T., et al. (2020) · 2006 14th Mediterranean Conference on Control and Automation · doi:10.1109/med48518.2020.9183266 · Cited by 12
“We have obtained this parameter by learning on the recorded data with worker moving in the simulated warehouse without robots and minimizing the average displacement error [24].”
MethodMulti object tracking: a survey
Bouraya, S., & Belangour, A. (2021) · International Conference on Digital Image Processing · doi:10.1117/12.2602901 · Cited by 12
Bartoli, F., Lisanti, G., Seidenari, L., Karaman, S., & Bimbo, A. D. (2015) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2015.7301279 · Cited by 13
“In order to assess the difficulties of this dataset we have also performed some tests exploiting seven representative state-of-the-art pedestrian detectors.”
MethodPatient–Robot Co-Navigation of Crowded Hospital Environments
Kodur, K., & Kyrarini, M. (2023) · Applied Sciences · doi:10.3390/app13074576 · Cited by 12
TrajPRed: Trajectory Prediction With Region-Based Relation Learning
Zhou, C., AlRegib, G., Parchami, A., et al. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3381843 · Cited by 11
“In this section, we empirically evaluate our prediction framework on the ETH[38]-UCY[39] dataset, and Stanford Drone Dataset (SDD)[40].”
MethodReview of the application of neural network approaches in pedestrian dynamics studies
Huang, S., Wei, R., Lian, L., et al. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e30659 · Cited by 11
“There are several publicly available pedestrian trajectory datasets: New York Grand Central (GC) [53], the ETH Dataset [54], UCY Dataset [55], CUHK Crowd Dataset [56], MIT Traffic Dataset [57], Town Center Dataset [58], and Stanford Drone Dataset (SDD) [59], shown as Fig.”
Towards Reliable Identification and Tracking of Drones Within a Swarm
Kumari, N., Lee, K., Barca, J. C., & Ranaweera, C. (2024) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-024-02115-1 · Cited by 11
“The artifacts may be tracked for example, swarms of drones performing target search [42, 43], recognising swarm formations [44, 45] post-disaster [46], swarm of drones in military operations (such as Iraq and Syria) [47], street pedestrians [48, 49], road sports automobiles on the court etc).”
Goal-Oriented Pedestrian Trajectory Prediction Considering Spatial–Temporal Interactions
Guo, H., Liu, Y., Meng, Q., et al. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3472774 · Cited by 11
“The proposed method is evaluated on the ETH [35] and UCY [36] datasets.”
MethodTracking Multiple Resolvable Group Targets With Coordinated Motion via Labeled Random Finite Sets
Wu, Q., Sun, J., Yang, B., et al. (2025) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2025.3539605 · Cited by 11
“…integrating group motion models into predictions to enhance tracking performance, such as the circular formation model [23], the leader-follower model (LFM) [24], [25], the social force model [26], [27], the stochastic differential equations (SDEs) model [28], [29] and so on [30], [31], [32], [33].”
Roadside Sensor Systems for Vulnerable Road User Protection: A Review of Methods and Applications
Zhang, T. T., Cheng, L., Bang, T., Guo, L., Hajij, M., Cao, S., Harris, A., & Sartipi, M. (2025) · IEEE Access · doi:10.1109/access.2025.3558174 · Cited by 15
“Using these coordinate systems to encode positions is particularly effective in static scenes, such as surveillance or drone footage [95], [96], where the sensor’s origin remains fixed.”
SocialMOIF: Multi-Order Intention Fusion for Pedestrian Trajectory Prediction
Chen, K., Zhao, X., Huang, Y., et al. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.02092 · Cited by 11
“ETH [32] and UCY [18] datasets encompass over 1,600 pedestrian trajectories, with positions annotated at 0.4s intervals.”
A multi-sensor cognitive approach for active security monitoring of abnormal overcrowding situations
Chiappino, S., Morerio, P., Marcenaro, L., et al. (2012) · CINECA IRIS Institutial Research Information System (University of Genoa) · Cited by 11
Tracking Millions of Humans in Crowded Spaces
Alahi, A., Ramanathan, V., & Fei-Fei, L. (2017) · Elsevier eBooks · doi:10.1016/b978-0-12-809276-7.00007-2 · Cited by 11
Sánchez, C. M., Zella, M., Capitán, J., & Marrón, P. J. (2022) · Sensors · doi:10.3390/s22031191 · Cited by 12
“[88] considered a multi-target prediction environment, in which people walk in crowded environments.”
BackgroundSceNDD: A Scenario-based Naturalistic Driving Dataset
Prabu, A., Ranjan, N., Li, L., et al. (2022) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc55140.2022.9921953 · Cited by 11
“Datasets such as the Stanford Drone Dataset [10], NGSIM [11], ETH [12], UCY [13], Town Center [14] are popular in motion”
MethodGame theoretical trajectory planning enhances social acceptability of robots by humans
Galati, G., Primatesta, S., Grammatico, S., et al. (2022) · Scientific Reports · doi:10.1038/s41598-022-25438-1 · Cited by 11
“The proposed game-theoretical human motion model is validated by conducting a qualitative comparison between generated trajectories and human ones, observed in open-source surveillance videos 37,38 .”
MethodSynchronous Bi-directional Pedestrian Trajectory Prediction with Error Compensation
Xie, C., Li, Y., Liang, R., et al. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-26351-4_42 · Cited by 11
Non-Autoregressive Sparse Transformer Networks for Pedestrian Trajectory Prediction
Liu, D., Li, Q., Li, S., Kong, J., & Qi, M. (2023) · Applied Sciences · doi:10.3390/app13053296 · Cited by 11
SG-LSTM: Social Group LSTM for Robot Navigation Through Dense Crowds
Bhaskara, R., Chiu, M., & Bera, A. (2023) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros55552.2023.10341954 · Cited by 11
“…social pooling • SG-LSTM (Ours): LSTM with social pooling and grouping And, we used the following metrics to measure the accuracy: • Average Displacement Error : We calculate the mean squared error for all the predicted points in trajectories for every individual present in the scene as in [7].”
MethodAnalyzing Groups: A Social Signaling Perspective
Bazzani, L., Cristani, M., Paggetti, G., et al. (2012) · Studies in computational intelligence · doi:10.1007/978-3-642-28598-1_9 · Cited by 11
“the work that deal with groups assumes that people are simple points on a plane [23, 32, 78, 53, 48, 84, 39] that in some cases may obey to physical laws of attraction and repulsion [54, 66].”
BackgroundIntention-Aware Multiple Pedestrian Tracking
Madrigal, F., Hayet, J.-B., & Lerasle, F. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.706 · Cited by 11
“In [9], the targets interact in such a way that they choose a collision-free trajectory.”
BackgroundInserting virtual pedestrians into pedestrian groups video with behavior consistency
Ren, Z., Gai, W., Zhong, F., et al. (2013) · The Visual Computer · doi:10.1007/s00371-013-0853-x · Cited by 11
“[6] propose a model designed for walking people with short-term prediction in mind, which improving the tracking performance in busy scenarios.”
A Real-Time Algorithm for Small Group Detection in Medium Density Crowds
Shao, J., Dong, N., & Zhao, Q. (2018) · Pattern Recognition and Image Analysis · doi:10.1134/s1054661818020074 · Cited by 11
“Some others [9] concentrated on studies on low density crowds, implementing experiments on datasets, such as the BIWI Walking Pedes-trian dataset [11].”
Hu, W., Shi, X., Zhou, Z., Xing, J., Ling, H., & Maybank, S. J. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01231-y · Cited by 12
“The classic social force model (Helbing and Molnar 1995) used in pedestrian tracking (Luber et al. 2010; Pellegrini et al. 2009; Scovanner and Tappen 2009) defines a series of social forces for an object to ensure collision avoidance and a desired direction for the destination.”
BackgroundMethodSocial-STAGE: Spatio-Temporal Multi-Modal Future Trajectory Forecast
Malla, S., Choi, C., & Dariush, B. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561582 · Cited by 11
“We evaluate 5 different scenarios on publicly available ETH [28]-UCY [29] datasets that have been widely used in”
MethodInfluentialOnline Multiple Athlete Tracking with Pose-Based Long-Term Temporal Dependencies
Kong, L., Zhu, M., Ran, N., Liu, Q., & He, R. (2020) · Sensors · doi:10.3390/s21010197 · Cited by 11
Detection of groups in crowd considering their activity state
Nakamura, K., Ono, T., & Babaguchi, N. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7899646 · Cited by 11
“As the datasets for the experiment, we employed “seq eth (eth)” and “seq hotel (hotel)” in BIWI Walking Pedestrians dataset [15] as well as “student003 (stu003)” in Crowds-ByExamples dataset [16], which were also used in the previous work [12].”
MethodData Association for Multi-Object Visual Tracking
Betke, M., & Wu, Z. (2016) · Synthesis lectures on computer vision · doi:10.2200/s00726ed1v01y201608cov009 · Cited by 11
Dong, B., Liu, H., Bai, Y., Lin, J., Xu, Z., Xu, X., & Kong, Q. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.16273 · Cited by 20
“The ETH[16] and UCY[11] datasets for pedestrian trajectory prediction only, include 5 scenes in total, including ETH, HOTEL, ZARA1, ZARA2, and UNIV.”
MethodGong, H., Li, Z., Lu, C., et al. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3328294 · Cited by 10
“3) Metrics: Average Displacement Error (ADE) [38], and Final Displacement Error (FDE) [39] are two metrics commonly used in trajectory prediction problems to accurately evaluate the model performance.”
BackgroundMeng, D., Zhao, G., & Yan, F. (2024) · Information · doi:10.3390/info15060341 · Cited by 10
“Social-STGMLP is trained on the ETH [21] and UCY [22] datasets.”
MethodTrajectory Prediction for Robot Navigation using Flow-Guided Markov Neural Operator
Bhaskara, R., Viswanath, H., & Bera, A. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10611154 · Cited by 10
“3) Comprehensive Evaluation: The model’s performance is evaluated on multiple datasets, including ETH and HOTEL [26], ZARA1, ZARA2, UCY [27], and RGB-D pedestrian [71].”
MethodMixed strategy Nash equilibrium for crowd navigation
Sun, M., Baldini, F., Hughes, K., Trautman, P., & Murphey, T. D. (2024) · The International Journal of Robotics Research · doi:10.1177/02783649241302342 · Cited by 12
“We prove the convergence of the sequence through the monotone convergence theorem, where we show that the sequence monotonically decreases on a function with finite lower bound.”
Adaptive graph transformer with future interaction modeling for multi-agent trajectory prediction
Chen, X., Wang, J., Deng, F., & Li, Z. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113363 · Cited by 10
Tracking by Prediction: A Deep Generative Model for Mutli-person Localisation and Tracking
Fernando, T., Denman, S., Sridharan, S., & Fookes, C. (2018) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2018.00128 · Cited by 68
“[46] 32.”
Analyzing Multiagent Interactions in Traffic Scenes via Topological Braids
Mavrogiannis, C., DeCastro, J., & Srinivasa, S. S. (2022) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra46639.2022.9812118 · Cited by 10
“as pedestrian tracking [31] or sports analysis [39].”
BackgroundScenario-based Evaluation of Prediction Models for Automated Vehicles
Sánchez, M. M., Elfring, J., Silvaş, E., & van de Molengraft, R. (2022) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc55140.2022.9922603 · Cited by 11
“a) A Constant Velocity (CV) model, which is often used as a baseline [22].”
MethodYang, Z., Pang, C., & Zeng, X. (2023) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi12020077 · Cited by 10
InstMove: Instance Motion for Object-centric Video Segmentation
Liu, Q., Wu, J., Jiang, Y., Bai, X., Yuille, A., & Bai, S. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.00614 · Cited by 12
“In crowded environments, pedestrian motion becomes more complex, prompting researchers to utilize the more expressive Social Force Mode [1, 48, 54, 55, 75].”
BackgroundIE-GAN: a data-driven crowd simulation method via generative adversarial networks
Lin, X., Liang, Y., Zhang, Y., et al. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17346-x · Cited by 10
“Our model is mainly performed on two common pedestrian datasets, ETH and UCY [12, 40], to verify the generation effect of IE-GAN.”
SemNav-HRO: A target-driven semantic navigation strategy with human–robot–object ternary fusion
Chen, B., Lu, S., Zhong, P., Cui, Y., Liang, Y., & Wang, J. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107370 · Cited by 10
Leveraging single for multi-target tracking using a novel trajectory overlap affinity measure
Manén, S., Timofte, R., Dai, D., & Gool, L. V. (2016) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2016.7477566 · Cited by 10
“2) Interaction models assume a fair amount of prior knowledge about the targets and/or large annotated training materials, which limits their applicability [31].”
BackgroundOnline Multi-object Tracking Based on Hierarchical Association Framework
Ju, J., Kim, D., Ku, B., Ko, H., & Han, D. K. (2016) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2016.161 · Cited by 10
The Analysis of High Density Crowds in Videos
Rodriguez, M. D., Šivic, J., & Ivan Laptev, A. (2017) · Elsevier eBooks · doi:10.1016/b978-0-12-809276-7.00006-0 · Cited by 10
Metadata
- Authors
- Pellegrini, S., Ess, A., Schindler, K., & van Gool, L.
- Affiliations
- ETH Zurich (CH)
- Technische Universität Darmstadt (DE)
- Venue
- 2009 IEEE 12th International Conference on Computer Vision
- Publisher
- IEEE
- Published
- 2009-09
- Pages
- 261-268
- OpenAlex
- W2532516272
- Semantic Scholar
- 7d7a98a81715e2d3747071722b2deeed8937d122
- Cited by
- 1,683
- References
- 25
- Citations by year
- 2026: 85 · 2025: 185 · 2024: 192 · 2023: 163 · 2022: 153 · 2021: 156 · 2020: 141 · 2019: 100 · 2018: 66 · 2017: 68 · 2016: 86 · 2015: 69 · 2014: 75 · 2013: 54 · 2012: 42
- FWCI
- 24.50
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE/CVF International Conference on Computer Vision (ICCV) (publishedVersion)
- Lirias (KU Leuven) (submittedVersion)
- Funders
- Else Kröner-Fresenius-Stiftung
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Social Sciences › Safety Research › Evacuation and Crowd Dynamics
- Keywords
- multi-target tracking, social behavior modeling, crowd simulation, collision avoidance, occlusion handling, vehicle-mounted camera, bird's-eye view, data association, human trajectory prediction, social interaction
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
