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
Research works printed but not linked: 1
- schadschneider, 2001, Cellular automaton approach to pedestrian dynamics—theory, PED
Cited by
Sorted by citations
Items 901–1,000 of 1,000
Progressive Contextual Trajectory Prediction With Adaptive Gating and Fuzzy Logic Integration
Sharma, N., Dhiman, C., & Indu, S. (2024) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2024.3391898 · Cited by 7
“The proposed method shows remarkable performance on both First-person-view(FPV) datasets like JAAD/PIE [10], [14] and Bird’s eye view(BEV) datasets like ETH/UCY [15], [16] surpassing baselines and state-of-the-art trajectory prediction methods.”
Evaluation of LSTM GAN for Trajectory Prediction in Connected and Autonomous Vehicles
Mistry, V., Vaidya, B., & Mouftah, H. T. (2024) · International Wireless Communications and Mobile Computing Conference (IWCMC) · doi:10.1109/iwcmc61514.2024.10592580 · Cited by 7
“The LSTM GAN model’s performance in trajectory prediction is evaluated using two standard metrics: Average Displacement Error (ADE) and Final Displacement Error (FDE) [8].”
MethodChen, W.-H., Wu, J.-C., Davydov, Y., Yeh, W.-C., & Lin, Y. (2024) · Sensors · doi:10.3390/s24155066 · Cited by 7
“With the steady development of deep learning technology and the release of public datasets [6–8], the research on pedestrian trajectory prediction has been rapidly advancing.”
BackgroundResultMATRIX: Multi-Agent Trajectory Generation with Diverse Contexts
Xu, Z., Zhou, R. H., Yin, Y., Gao, H., Tomizuka, M., & Li, J. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610944 · Cited by 7
“A. Experiment Setup 1) Data Preparation: MATRIX is trained on two widely used datasets: the ETH dataset [58], with subsets named ETH and HOTEL, and the UCY dataset [59], with subsets named ZARA1, ZARA2, and UNIV.”
MethodSocial Zone as a Barrier Function for Socially-Compliant Robot Navigation
Jang, J., & Ghaffari, M. (2024) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2025.01.173 · Cited by 7
Li, R., Katsigiannis, S., Kim, T., & Shum, H. P. H. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2025.3545268 · Cited by 10
“We evaluate BP-SGCN on multiple benchmark datasets, including the SDD [30], Argoverse 1 [34], ETH [20] and UCY [21], and the pedestrian-only version of SDD [35].”
Online Social Behavior Modeling for Multi-target Tracking
Zhang, S., Das, A., Ding, C., et al. (2013) · IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2013.113 · Cited by 7
“In many application domains, the tracker needs to work without the advantage of having seen videos captured under similar circumstances from which the SBMs can be learned a priori (as in [15, 21, 26]).”
BackgroundCodec independent region of interest video coding using a joint pre- and postprocessing framework
Meuel, H., Munderloh, M., Kluger, F., et al. (2016) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2016.7552876 · Cited by 7
“2 dB [20] Video conferencing sequence, 100 frames, PSNR ≈ 43.”
MethodEncoding Motion Cues for Pedestrian Path Prediction in Dense Crowd Scenarios
Li, Y., Mekhalfi, M. L., Rahhal, M. M. A., et al. (2017) · IEEE Access · doi:10.1109/access.2017.2772029 · Cited by 7
“As for comparison, we implemented three well-known reference methods for crowd analysis, namely Social Force Dynamic (SF) [22], Constant Velocity (CV) [26], and Constant Acceleration (CA) [26].”
MethodInfluentialDeepGoal: Learning to drive with driving intention from human control demonstration
Ma, H., Wang, Y., Xiong, R., Kodagoda, S., & Tang, L. (2020) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2020.103477 · Cited by 7
A Toolkit to Generate Social Navigation Datasets
Baghel, R., Kapoor, A., Bachiller, P., Jorvekar, R. R., Rodriguez-Criado, D., & Manso, L. J. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-62579-5_13 · Cited by 7
“The video datasets ETH [8] and UCY [9] were used in [10] to train a social attention model.”
BackgroundMethodCNN, Segmentation or Semantic Embeddings: Evaluating Scene Context for Trajectory Prediction
Syed, A., & Morris, B. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-64559-5_56 · Cited by 7
Indoor Future Person Localization from an Egocentric Wearable Camera
Qiu, J., Lo, F. P.-W., Gu, X., Sun, Y., Jiang, S., & Lo, B. P. L. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9635868 · Cited by 12
“Future person location and movement trajectory prediction have been an active research area, but predictions in most studies have only been carried out in videos captured from birds-eye or oblique views [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17].”
BackgroundSubaih, R., Maree, M., Tordeux, A., & Chraibi, M. (2022) · Applied Sciences · doi:10.3390/app12157563 · Cited by 6
“[18], takes into account both simple scene information in the form of destinations or desired directions and interactions between different pedestrians.”
BackgroundMethodAutomatic Identification of Human Subgroups in Time-Dependent Pedestrian Flow Networks
Wu, W., Yi, W., Li, J., et al. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3262975 · Cited by 7
Learning social navigation from demonstrations with conditional neural processes
Yıldırım, Y., & Uğur, E. (2022) · Interaction Studies Social Behaviour and Communication in Biological and Artificial Systems · doi:10.1075/is.22018.yil · Cited by 7
“In the following, we explain these modules in detail.”
BackgroundAbnormalities detection and tracking of crowd using kinematics approach in a video stream
Manu, Y. M., & Ravikumar, G. K. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-19146-3 · Cited by 7
“In papers [21, 22], optimization methods such as the gradient-based Newton algorithm and genetic algorithm is used in order to obtain the proper parameter values for motion model learning.”
On Pairwise Cost for Multi-Object Network Flow Tracking
Chari, V., Lacoste-Julien, S., Ivan Laptev, A., et al. (2015) · HAL (Le Centre pour la Communication Scientifique Directe) · Cited by 7
Hierarchically-constrained optical flow
Kennedy, R., & Taylor, C. J. (2015) · Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2015.7298955 · Cited by 11
“2 do in fact occur in real images we evaluated our algorithm on the Hotel sequence from the BIWI Walking Pedestrians dataset [24], which is a video of a city sidewalk taken from an overhead camera.”
MethodFilling in the blanks: reconstructing microscopic crowd motion from multiple disparate noisy sensors
Yoon, S., Kapadia, M., Sahu, P., et al. (2016) · IEEE Winter Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw.2016.7470118 · Cited by 7
Unsupervised camera localization in crowded spaces
Alahi, A., Wilson, J., Fei-Fei, L., & Savarese, S. (2017) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2017.7989311 · Cited by 7
Khan, G., Tariq, Z., Hussain, J., et al. (2019) · International Conference on Communication Technologies (ComTech) · doi:10.1109/comtech.2019.8737838 · Cited by 7
“Training and testing are performed on BIWI [30] and CBE datasets [31].”
MethodInfluentialThe Role of Data-Driven Priors in Multi-Agent Crowd Trajectory Estimation
Qiao, G., Yoon, S., Kapadia, M., et al. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11593 · Cited by 7
“Similar to these model-based approaches (Pellegrini et al. 2009; Bera, Kim, and Manocha 2015), our approach also relies on a synthetic simulator to provide motion priors for crowd trajectory estimation, but it utilizes the prior as a term within a modified global optimization framework of (Yoon et…”
MethodApproaches to Video Real time Multi-Object Tracking and Object Detection: A survey
Bouraya, S., & Belangour, A. (2021) · International symposium on image and signal processing and analysis/ISPA ... · doi:10.1109/ispa52656.2021.9552095 · Cited by 7
Predicting the Future Motion of Divers for Enhanced Underwater Human-Robot Collaboration
Agarwal, T., Fulton, M., & Sattar, J. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636374 · Cited by 7
“Popular datasets used for motion prediction, such as the ETH [29], UCY, and Zara datasets [30], consist of videos of pedestrians recorded from a stationary camera from a rooftop or window.”
MethodSSP: Single Shot Future Trajectory Prediction
Dwivedi, I., Malla, S., Dariush, B., & Choi, C. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9340754 · Cited by 6
“To extract such features, we annotate a segmentation map for each scene in the ETH, UCY, and SDD datasets.”
BackgroundInfluentialA Game-Theoretic Approach to Multi-Pedestrian Activity ForecastingPreprint
Ma, W.-C., Huang, D.-A., Lee, N., & Kitani, K. (2016) · arXiv (Cornell University) · Cited by 7
Xue, H., Huynh, D. Q., & Reynolds, M. A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.05507 · Cited by 7
“Both SGSG and SR-LSTM 2 attain the smallest average ADE of 0.45m.”
BackgroundMethodInfluentialMBT3D: Deep learning based multi-object tracker for bumblebee 3D flight path estimation
Stiemer, L., Thoma, A., & Braun, C. (2023) · PLoS ONE · doi:10.1371/journal.pone.0291415 · Cited by 6
“These trackers often use Social Force Models for motion prediction [10, 11].”
MethodImproving trajectory prediction in dynamic multi-agent environment by dropping waypoints
Chib, P. S., & Singh, P. P. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112240 · Cited by 6
FuseFormer: A Manifold Metric Fusing Attention for Pedestrian Trajectory Prediction
Zou, Y., Ko, K., Yang, J., et al. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3565990 · Cited by 6
“A. Datasets and Evaluation 1) Dataset: We evaluated our methodology on the following three public datasets: ETH [59], UCY [60], and Stanford Drone Dataset (SDD) [61], NBA SportVU(NBA)dataset [27].”
Adapting to Observation Length of Trajectory Prediction via Contrastive Learning
Qiu, R., Gong, J., Zhang, X., 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/cvpr52734.2025.00161 · Cited by 6
“Early attempts at human trajectory prediction relied on manually designed features based on prior knowledge [10, 20, 24].”
Modelling Diverse Interactions and Multimodality for Pedestrian Trajectory Prediction
Wang, R., Hu, Z., Yu, J., et al. (2025) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2025.125363 · Cited by 6
Yan, D., Ding, G., Huang, K., et al. (2024) · The Visual Computer · doi:10.1007/s00371-024-03385-4 · Cited by 6
“The ETH dataset [55] consists of pedestrian data from two scenes, ETH and Hotel.”
DTDNet: Dynamic Target Driven Network for pedestrian trajectory prediction
Liu, S., Sun, J., Yao, P., et al. (2024) · Frontiers in Neuroscience · doi:10.3389/fnins.2024.1346374 · Cited by 6
Pedestrian Trajectory Prediction Using Dynamics-based Deep Learning
Wang, H., Zhi, W., Batista, G., & Chandra, R. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10609993 · Cited by 9
“The dataset is divided into two parts: the ETH dataset [38] and the UCY dataset [39].”
MethodFast and Accurate Multi-Agent Trajectory Prediction for Crowded Unknown Scenes
Tao, X., Li, H., Liang, B., et al. (2024) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2024.3466272 · Cited by 6
“To evaluate the proposed method, we use pedestrian walking datasets of ETH, HOTEL, UNIV, ZARA1 and ZARA2 in [5], [36] to test the proposed algorithm.”
MethodDySeT: A Dynamic Masked Self-distillation Approach for Robust Trajectory Prediction
Pourkeshavarz, M., Zhang, J., & Rasouli, A. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72646-0_19 · Cited by 6
Socially Acceptable Bipedal Robot Navigation via Social Zonotope Network Model Predictive Control
Shamsah, A., Agarwal, K., Katta, N., et al. (2024) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2024.3519012 · Cited by 6
“The core assumption underlying our framework is that the paths recorded in human crowd datasets [70], [71] represent socially acceptable trajectories.”
Chen, B., Guo, R., Zhang, Q., Zhao, Y., Wang, X., & Zhu, Z. (2025) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2025.3549150 · Cited by 8
“We evaluate our proposed crowd simulation framework on two public large-scale real-world datasets, including the GC dataset [58] and the UCY dataset [59].”
Beyond Pedestrians: A Hybrid Approach of Tracking Multiple Articulating Humans
Wang, W., Nevatia, R., & Yang, B. (2015) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2015.25 · Cited by 6
A novel model based on deep learning for Pedestrian detection and Trajectory prediction
Shi, K., Zhu, Y., & Pan, H. (2019) · IEEE Joint International Information Technology and Artificial Intelligence Conference (ITAIC) · doi:10.1109/itaic.2019.8785741 · Cited by 6
“(8) (9) The average displacement error ADE is used to characterize the accuracy of the prediction sequence at all prediction times t, and the final displacement error FDE is used to characterize the accuracy of the prediction sequence accumulation at the last moment tpred.”
MethodInfluentialMeta-path Analysis on Spatio-Temporal Graphs for Pedestrian Trajectory Prediction
Hasan, A., Sriram, P., & Driggs-Campbell, K. (2022) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra46639.2022.9811632 · Cited by 7
“However, this is not to say that the model does not perform well in the absence of data from neighbouring pedestrians, as the model still outperforms both baselines on the scenes from the ETH dataset.”
MethodInfluentialStructure entropy minimization-based dynamic social interaction modeling for trajectory prediction
Jin, Y., Yang, S., Lv, W., et al. (2022) · Information Sciences · doi:10.1016/j.ins.2022.10.024 · Cited by 6
Effective Multi-Object Tracking via Global Object Models and Object Constraint Learning
Yoo, Y., Lee, S.-H., & Bae, S. (2022) · Sensors · doi:10.3390/s22207943 · Cited by 9
“To learn our global relation motion model, we exploit ETH [54], UCY [60], and MOT15 datasets.”
MethodCross-domain Trajectory Prediction with CTP-Net
Huang, P., Cui, Z., Li, J., Gao, S.-H., Hu, B., & Fang, Y.-Y. (2021) · CAAI International Conference on Artificial Intelligence · doi:10.1007/978-3-031-20497-5_7 · Cited by 4
“Source: UCY-zara02; Target: ETH-univ.”
InfluentialTrajectory Prediction of Moving Workers for Autonomous Mobile Robots on the Shop Floor
Löcklin, A., Artelt, M., Ruppert, T., Vietz, H., Jazdi, N., & Weyrich, M. (2022) · IEEE International Conference on Emerging Technologies and Factory Automation (ETFA) · doi:10.1109/etfa52439.2022.9921493 · Cited by 6
“The trajectory prediction algorithm is optimized for open spaces without obstacles and trained on corresponding data without obstacle structures, specifically the UCY [35] and ETH [36] datasets.”
MethodYang, X., Fan, J., & Xing, S. (2023) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-023-01889-4 · Cited by 6
“In this section, we evaluate our IST-PTEPN on two public datasets: ETH [25] & UCY [26], and adopt the Average Displacement Error (ADE) and Final Displacement Error (FDE) as the criteria in our trajectory prediction.”
MethodInfluentialTTC-SLSTM: Human Trajectory Prediction Using Time-to-Collision Interaction Energy
Dang, H. S., Korbmacher, R., Tordeux, A., et al. (2023) · International Conference on Knowledge and Systems Engineering/Proceedings · doi:10.1109/kse59128.2023.10299443 · Cited by 6
“Two publicly available datasets are used to train the neural network and evaluate its predictions, which include ETH [17] and UCY [18].”
MethodOptimizing Algorithms from Pairwise User Preferences
Keselman, L., Shih, K., Hebert, M., et al. (2023) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros55552.2023.10342081 · Cited by 6
“We study social navigation in the context of the recent benchmark SocNavBench [32], a simulation environment that contains pedestrian interaction scenarios sourced from the ETH and UCY datasets [33], [34].”
MethodAlmeida, T., Rudenko, A., Schreiter, T., Zhu, Y., Maestro, E. G., Morillo-Mendez, L., Kucner, T. P., Mozos, Ó. M., Magnusson, M., Palmieri, L., et al. (2023) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw60793.2023.00234 · Cited by 6
“While CVM may serve as a relevant baseline in ETH/UCY benchmarks [17, 13], the TH ¨OR-Magni dataset presents signifi-cantly more challenging and complex data, where simplistic baselines like CVM face substantial limitations.”
BackgroundReinforcement Learning Based User-Specific Shared Control Navigation in Crowds
Zhang, B., Holloway, C., & Carlson, T. (2023) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc53992.2023.10394139 · Cited by 6
Yang, X., Fan, J., Wang, X., et al. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-09784-x · Cited by 6
“6 Architecture of the CVAE conditional generation model CSGAT-Net is tested on two publicly available datasets: ETH [32] and UCY [33].”
TrajFlow: Learning Distributions over Trajectories for Human Behavior Prediction
Mészáros, A., Schumann, J. F., Alonso–Mora, J., Zgonnikov, A., & Kober, J. (2024) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv55156.2024.10588386 · Cited by 6
“We validated our approach on a synthetic dataset for which we know the underlying distribution, as well as on an augmented version of the multi-modal Forking Paths dataset [25], and several popular real-world datasets (ETH/UCY [26], [27], rounD [28], and nuScenes [29]).”
MethodDATraj: A Dynamic Graph Attention Based Model for Social-Aware Pedestrian Trajectory Prediction
Si, Z., Ye, P., Xiong, G., Lv, Y., & Zhu, F. (2024) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv55156.2024.10588693 · Cited by 6
“Our proposed model's performance is trained and evaluated using the ETH/UCY [28],[29].”
MethodChiappino, S., Morerio, P., Marcenaro, L., et al. (2013) · CINECA IRIS Institutial Research Information System (University of Genoa) · Cited by 6
Detection of pedestrian crossing road using action classification model
Hariyono, J., & Jo, K.-H. (2015) · IEEE International Conference on Advanced Intelligent Mechatronics (AIM) · doi:10.1109/aim.2015.7222502 · Cited by 6
Hierarchical Group Structures in Multi-person Tracking
Yan, X., Cheriyadat, A., & Shah, S. K. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.386 · Cited by 6
“[10] accounts for social repulsion effects among pedestrians to predict motion paths using a social force model.”
MethodTracking People by Evolving Social Groups: An Approach with Social Network Perspective
Feng, L., & Bhanu, B. (2015) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2015.22 · Cited by 6
“Recent research efforts [14, 18, 19] have suggested that social groups can be used to improve tracking performance, especially in semi-crowded environments where individuals are still discernible while occlusions happen frequently.”
BackgroundYang, W. P., Song, Z. P., & Wu, X. (2012) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2012.6490988 · Cited by 6
“To verify the effectiveness of the evaluation function, we compared the performance of our method with the direct subtraction and the method in [17] on BIWI Walking Pedestrians dataset [18].”
MethodSocial force model aided robust particle PHD filter for multiple human tracking
Feng, P., Wang, W., Naqvi, S. M., et al. (2016) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2016.7472508 · Cited by 6
“Pedestrian behaviours have been studied from a crowd perspective, with macroscopic models for pedestrian density and velocity [14].”
BackgroundMethodInfluentialLearning pedestrian dynamics with kriging
Kawamoto, K., Tomura, Y., & Okamoto, K. (2016) · IEEE/ACIS International Conference on Computer and Information Science (ICIS) · doi:10.1109/icis.2016.7550877 · Cited by 6
“For the determination of the model parameters, computationally expensive methods such as evolutionary computation [5], [6] and particle filter [7], [8] are required.”
BackgroundMethodPedestrian group attributes detection in crowded scenes
Shuaibu, A. N., Malik, A. S., Faye, I., et al. (2017) · International Conference on Advanced Technologies for Signal and Image Processing · doi:10.1109/atsip.2017.8075584 · Cited by 6
“Most of the existing crowd analysis techniques focus more on a collection of individual [19, 20] or as an aggregated whole crowd [21, 6, 7].”
BackgroundModeling Trajectory-level Behaviors using Time Varying Pedestrian Movement Dynamics
Bera, A., Kim, S., & Manocha, D. (2018) · Collective Dynamics · doi:10.17815/cd.2018.15 · Cited by 6
“[39] use an energy function to build up a goal-directed short-term collision-avoidance motion model.”
MethodInteractive Multiple-Target Tracking via Labeled Multi-Bernoulli Filters
Gostar, A. K., Rathnayake, T., Fu, C., et al. (2019) · International Conference on Control, Automation and Information Sciences (ICCAIS) · doi:10.1109/iccais46528.2019.9074567 · Cited by 6
“In social force models each target is considered to be dependent on other targets which determine its velocity, and acceleration [27], [28], [29], [30], [31], [32].”
BackgroundGraph-based topic models for trajectory clustering in crowd videos
Alghamdi, M., & Gotoh, Y. (2020) · Machine Vision and Applications · doi:10.1007/s00138-020-01092-3 · Cited by 6
“Based on the social force model [23], Pellegrini et al. [21] proposed a linear trajectory avoidance (LTA) model to predict the optimal path for individuals that prevents collisions with each other and the obstacles.”
MethodData-driven entropic spatially inhomogeneous evolutionary games
Bonafini, M., Fornasier, M., & Schmitzer, B. (2022) · European Journal of Applied Mathematics · doi:10.1017/s0956792522000043 · Cited by 7
“Related datasets and models in the computer vision community can be found, for instance, in [59, 76, 62].”
BackgroundMap Feature Based Trajectory Prediction with Multi-class Traffic Participants
Zuo, Z., Zhang, X., & Wang, Y. (2021) · Chinese Control Conference (CCC) · doi:10.23919/ccc52363.2021.9549491 · Cited by 6
“The former [2, 14] depends on manually designed rules and features to learn the movement patterns and interactions of traffic participants.”
BackgroundSohn, S. S., Lee, M., Moon, S., et al. (2022) · Computers & Graphics · doi:10.1016/j.cag.2022.05.010 · Cited by 6
SteerFit: Automated Parameter Fitting for Steering Algorithms
Guy, S. J., Karamouzas, I., Lin, M. C., et al. (2016) · CRC Press eBooks · doi:10.1201/9781315370071-12 · Cited by 6
A Location Privacy Preserving Method Based on Sensitive Diversity for LBS
Zhou, C., Ma, C., Yang, S., et al. (2014) · Lecture notes in computer science · doi:10.1007/978-3-662-44917-2_34 · Cited by 6
“Staying at a sensitive place causes a semantic association with a user, continuous sensitive places lead to deep-going leakages[10,11].”
BackgroundRoadTrack: Tracking Road Agents in Dense and Heterogeneous Environments.Preprint
Chandra, R., Bhattacharya, U., Randhavane, T., Bera, A., & Manocha, D. (2019) · arXiv (Cornell University) · Cited by 6
End-to-end Recurrent Multi-Object Tracking and Trajectory Prediction with Relational ReasoningPreprint
Fuchs, F. B., Kosiorek, A. R., Sun, L., Jones, Ō. P., & Posner, I. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.12887 · Cited by 6
“Initial research modelled social forces using hand-crafted features [21, 22, 23, 24] or MDP-based motion transition models [25], while more recent approaches learn from context information, e.”
MethodAutonomous Social Distancing in Urban Environments using a Quadruped RobotPreprint
Fan, T., Chen, Z., Zhao, X., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.08889 · Cited by 6
“Pellegrini et al. developed Linear Trajectory Avoid-ance (LTA) to improve the accuracy of motion prediction [13].”
BackgroundEvidential Sparsification of Multimodal Latent Spaces in Conditional Variational AutoencodersPreprint
Itkina, M., Ivanovic, B., Senanayake, R., Kochenderfer, M. J., & Pavone, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.09164 · Cited by 9
“We evaluate Trajectron++’s performance with different probability filtering schemes on 203 randomly-sampled examples from the test set of the ETH pedestrian dataset [35], consisting of real world human trajectories with rich interaction scenarios.”
MethodSFU-store-nav: A multimodal dataset for indoor human navigation
Zhang, Z., Rhim, J., TaherAhmadi, M., et al. (2020) · Data in Brief · doi:10.1016/j.dib.2020.106539 · Cited by 6
Context-aware pedestrian motion prediction
Haddad, S. (2021) · doi:10.32657/10356/151544 · Cited by 6
AmeliaTF: A Large Model and Dataset for Airport Surface Movement Forecasting
Navarro, I., Ortega-Kral, P., Patrikar, J., Wang, H., Ye, Z., Park, J. H., Oh, J., & Scherer, S. (2024) · AIAA AVIATION Forum · doi:10.2514/6.2024-4251 · Cited by 6
End-to-end pedestrian trajectory prediction via Efficient Multi-modal Predictors
Wu, Q., Zhou, S., Wang, L., Shi, L., Dong, Y., & Hua, G. (2024) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2024.104107 · Cited by 5
A hypergraph-based dual-path multi-agent trajectory prediction model with topology inferring
Hu, Y., Chen, X., Zhou, Y., et al. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110799 · Cited by 5
Highly Accurate and Diverse Traffic Data: The DeepScenario Open 3D Dataset
Dhaouadi, O., Meier, J., Wahl, L., Kaiser, J., Scalerandi, L., Wandelburg, N., Zhou, Z., Berinpanathan, N., Banzhaf, H., & Cremers, D. (2025) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv64158.2025.11097484 · Cited by 7
“The infrastructure camera datasets, like NGSIM [15], BIWI [16], have fewer occlusions but suffer from low resolution [15] or lack of class diversity [16].”
Socially aware navigation for mobile robots: a survey on deep reinforcement learning approaches
Kabir, I. K., & Mysorewala, M. F. (2026) · Applied Intelligence · doi:10.1007/s10489-025-06982-5 · Cited by 9
“The ETH [145] and UCY [146] datasets are commonly used in SAN for trajectory prediction.”
Psalta, A., Tsironis, V., Karantzalos, K. G., et al. (2020) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc45102.2020.9294251 · Cited by 5
“Other methods, such as [2] [9], extract features in order to model social interactions (e.g. velocity, direction) and forecast future positions of pedestrians by minimizing an energy function based on those features.”
BackgroundMethodLiu, S., Liu, H., Wang, Y., Sun, J., & Mao, T. (2022) · Computational Intelligence and Neuroscience · doi:10.1155/2022/4192367 · Cited by 7
“We evaluate our method on three commonly used datasets, ETH [35], UCY [36], and a high-density pedestrian dataset, pedestrian walk path dataset [37], which is referred to as PEDWALK in the rest of the article.”
MethodTraining a Non-Cooperator to Identify Vulnerabilities and Improve Robustness for Robot Navigation
Qiu, Q., Zhang, X., Yao, S., Chen, Y., Chen, G., Hua, B., & Ji, J. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3284379 · Cited by 6
“These datasets include the ETH dataset [12], the Hotel dataset [12], the ZARA01 dataset [13].”
MethodActive Collision Avoidance System for E-Scooters in Pedestrian Environment
Yan, X., & Shen, D. (2024) · SAE International Journal of Advances and Current Practices in Mobility · doi:10.4271/2024-01-2555 · Cited by 5
Qiu, X., Wang, H., & Tan, X. (2024) · IEEE Systems Journal · doi:10.1109/jsyst.2024.3440472 · Cited by 5
“The dataset provides rich information on human movement and interactions in crowded spaces, with elaborate agent identity and spatial–temporal metadata annotations [42].”
BackgroundDou, W., & Lu, L. (2024) · Applied Sciences · doi:10.3390/app14209537 · Cited by 5
Hydrodynamics-Informed Neural Network for Simulating Dense Crowd Motion Patterns
Zhou, Y., Lai, P., Yu, J., et al. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681277 · Cited by 5
SPAMming Labels: Efficient Annotations for the Trackers of Tomorrow
Cetintas, O., Meinhardt, T., Brasó, G., & Leal-Taixé, L. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73254-6_22 · Cited by 5
“Modern trackers have explored performing (i) and (ii) jointly [7, 25, 45, 80], or providing solutions for (ii) relying on motion models [2, 3, 39, 49, 51,54,55,71,77], appearance models [13,15,37,48,53,56,72,78] or more complex techniques involving transformers [10,74].”
BackgroundTranSPORTmer: A Holistic Approach to Trajectory Understanding in Multi-agent Sports
Capellera, G., Ferraz, L., Rubio, A., Agudo, A., & Moreno-Noguer, F. (2024) · Lecture notes in computer science · doi:10.1007/978-981-96-0901-7_20 · Cited by 11
Mei, L., Fu, M., Wang, B., Jia, L., Yu, M., Zhang, Y., & Zhang, L. (2025) · Remote Sensing · doi:10.3390/rs17010154 · Cited by 5
“…FPS = 1 and obtains an input sequence of t p = 5 s and output of length t f = 30 s. ETH-UCY Dataset: Many previous methods such as [21,22] use several sub-datasets from ETH [53] and UCY [54] to train and evaluate their models with the “leave-one-out” strategy, which is called the ETH-UCY dataset.”
Intersection-Specific Trajectory Prediction for Road Users: A Review
Guo, X., Adl, M., Abdi, B., & Emadi, A. (2025) · IEEE Access · doi:10.1109/access.2025.3546325 · Cited by 5
SHINE: Social homology identification for navigation in crowded environments
Martínez, D., de Groot, O., Knoedler, L., et al. (2025) · The International Journal of Robotics Research · doi:10.1177/02783649251344639 · Cited by 5
“These methods prove their performance in public real-world datasets, such as the ETH/UCY dataset (Lerner et al. 2007) (Pellegrini et al. 2009), which have data of pedestrians positions in 5 different scenarios.”
InfluentialAnother Vertical View: A Hierarchical Network for Heterogeneous Trajectory Prediction via Spectrums
Xia, B., Wong, C., Xu, D., et al. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3590487 · Cited by 5
“It is an essential but challenging task, which can be widely applied to behavior analysis [1], robot navigation and planning [2], autonomous driving [3], tracking [4], detection [5], and many other tasks [6], [7].”
Conv-LSTM: Pedestrian Trajectory Prediction in Crowded Scenarios
Chen, K., Song, X., & Yu, H. (2019) · Communications in computer and information science · doi:10.1007/978-981-15-1078-6_3 · Cited by 5
Hug, R., Becker, S., Hübner, W., et al. (2021) · IEEE Access · doi:10.1109/access.2021.3082904 · Cited by 5
“Evaluations are conducted on scenes of the following standard benchmarking datasets: BIWI Walking Pedestrians ([13], abbrev.”
MethodPrediction of Pedestrian Risky Level for Intelligent Vehicles
Zhang, Z., Lu, C., Li, J., et al. (2020) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv47402.2020.9304707 · Cited by 5
“The collected trajectory is relatively short-duration, with an average duration of 21 frames (i.e., 3.23 seconds), compared to other absolute pedestrian trajectory datasets in non-traffic scenes [22].”
ResultInfluentialMultiple Object Tracking using YOLO-based Detector
Lin, S. D., Chang, T., & Chen, W. (2021) · Journal of Imaging Science and Technology · doi:10.2352/j.imagingsci.technol.2021.65.4.040401 · Cited by 5
Scene Transformer: A unified architecture for predicting multiple agent trajectories
Ngiam, J., Caine, B., Vasudevan, V., Zhang, Z., Chiang, H.-T. L., Ling, J., Roelofs, R., Bewley, A., Liu, C., Venugopal, A., et al. (2021) · arXiv · doi:10.48550/arxiv.2106.08417 · Cited by 84
“…on the interactive splits of the validation and test datasets (Ettinger et al., 2021) evaluated with scene-level joint metrics for t = 8 seconds. minSADE, minSFDE for k = 6 predictions (Alahi et al., 2016; Pellegrini et al., 2009); Miss Rate (MR) (Chang et al., 2019) within 2 meters of the target.”
BackgroundMethod
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
