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 201–300 of 1,000
Recent trends in crowd analysis: A review
Bendali-Braham, M., Weber, J., Forestier, G., Idoumghar, L., & Müller, P.-A. (2021) · Machine Learning with Applications · doi:10.1016/j.mlwa.2021.100023 · Cited by 86
Pedestrian Intention Prediction for Autonomous Vehicles: A Comprehensive Survey
Sharma, N., Dhiman, C., & Indu, S. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.07.085 · Cited by 84
THÖR: Human-Robot Navigation Data Collection and Accurate Motion Trajectories Dataset
Rudenko, A., Kucner, T. P., Swaminathan, C. S., et al. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.2965416 · Cited by 78
“The popular ETH dataset [2] is recorded outdoors in a pedestrian zone with a stationary camera facing downwards and manually annotated at 2 .”
BackgroundMethodInfluentialHow many Observations are Enough? Knowledge Distillation for Trajectory Forecasting
Monti, A., Porrello, A., Calderara, S., et al. (2022) · 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/cvpr52688.2022.00644 · Cited by 77
“Pi-oneering works take advantage of hand-crafted relations, energy-based features or rule-based models [3, 16, 26, 40, 44], which fail to adapt to scene changes and model complex crowd dynamics.”
(MP)2T: Multiple People Multiple Parts Tracker
Izadinia, H., Saleemi, I., Li, W., & Shah, M. A. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33783-3_8 · Cited by 78
“1 64 Berclaz [13] 62 78 78 62 Pellegrini [19] 70.”
Learning to predict trajectories of cooperatively navigating agents
Kretzschmar, H., Kuderer, M., & Burgard, W. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907442 · Cited by 76
“The second dataset [17] comprises 12 minutes of trajectories of pedestrians interacting in a hotel entrance, leading to 71 composite trajectories with three to five pedestrians each.”
BackgroundIntrovert: Human Trajectory Prediction via Conditional 3D Attention
Shafiee, N., Padır, T., & Elhamifar, E. (2021) · 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/cvpr46437.2021.01654 · Cited by 75
“By experiment on UCY [27] and ETH [35] datasets, we show that out method significantly improves the state of the art performance, reducing the average prediction error on 5 datasets from 0.”
MethodHSTA: A Hierarchical Spatio-Temporal Attention Model for Trajectory Prediction
Wu, Y., Chen, G., Li, Z., Zhang, L., Xiong, L., Liu, Z., & Knoll, A. (2021) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2021.3115018 · Cited by 80
“We first conduct experiments on two pedestrian datasets ETH and UCY, then adopt data pre-processing methods in [9], [46], [47].”
MethodOccupancy Flow Fields for Motion Forecasting in Autonomous Driving
Mahjourian, R., Kim, J., Chai, Y., Tan, M., Sapp, B., & Anguelov, D. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3151613 · Cited by 77
“This set may be predicted in a discriminative, feed-forward manner, via imitation learning [8], [12], [4], [13], [14], [15], [7], [16], [7], [5], [6].”
BackgroundPredicting Object Dynamics in Scenes
Fouhey, D. F., & Zitnick, C. L. (2014) · 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.2014.260 · Cited by 73
Multiple Trajectory Prediction of Moving Agents With Memory Augmented Networks
Marchetti, F., Becattini, F., Seidenari, L., & Bimbo, A. D. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3008558 · Cited by 78
“Several researchers have focused on trajectories of pedestrians [6], [7], [46], [47], [48], either regarded as individuals or crowds, also exploiting social behaviors and interactivity between individuals [6], [7], [46], [47], [49], [50].”
BackgroundTrajectory Forecasting Based on Prior-Aware Directed Graph Convolutional Neural Network
Su, Y., Du, J., Li, Y.-M., Li, X., Liang, R., Hua, Z., & Zhou, J. (2022) · IEEE transactions on intelligent transportation systems (Print) · doi:10.1109/tits.2022.3142248 · Cited by 71
“The true positions of each agent are recorded in every ten frames, as implemented in ETH [40] and UCY [41].”
MethodGLMP- realtime pedestrian path prediction using global and local movement patterns
Bera, A., Kim, S., Randhavane, T., Pratapa, S., & Manocha, D. (2016) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2016.7487768 · Cited by 73
A Deep Spatiotemporal Perspective for Understanding Crowd Behavior
Li, Y. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2834873 · Cited by 74
“In the second part of the experiment, we extend the values of K and K ′ to 8 and 12, respectively, for the ETH and UCY datasets, as in [37] and [36].”
BackgroundMethodInfluentialLooking to Relations for Future Trajectory Forecast
Choi, C., & Dariush, B. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00101 · Cited by 75
“In addition to the evaluation using SDD, we perform the experiments on the ETH [27] and UCY [21] dataset, comparing with S-LSTM [1] and SS-LSTM [44].”
MethodSocNavBench: A Grounded Simulation Testing Framework for Evaluating Social Navigation
Biswas, A., Wang, A., Silvera, G., Steinfeld, A. M., & Admoni, H. (2022) · ACM Transactions on Human-Robot Interaction · doi:10.1145/3476413 · Cited by 72
“The pre-recorded pedestrian data used in SocNavBench comes from the UCY [25] and ETH [37] pedestrian trajectory datasets, which are widely accepted and used by the community [2, 5, 7, 15, 44].”
MethodModeling, Evaluation, and Scale on Artificial Pedestrians
Martínez-Gil, F., Lozano, M., García‐Fernández, I., & Fernández, F. (2017) · ACM Computing Surveys · doi:10.1145/3117808 · Cited by 71
“Another work that uses video tracking to train a model of a crowd is Pellegrini et al. (2009).”
BackgroundAutonomous Social Distancing in Urban Environments Using a Quadruped Robot
Chen, Z., Fan, T., Zhao, X., et al. (2021) · IEEE Access · doi:10.1109/access.2021.3049426 · Cited by 70
“Pellegrini et al. developed Linear Trajectory Avoid-ance (LTA) to improve the accuracy of motion prediction [13].”
BackgroundLearning Scene-Independent Group Descriptors for Crowd Understanding
Shao, J., Loy, C. C., & Wang, X. (2017) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2016.2539878 · Cited by 72
“Most of the existing imagery-based crowd analysis methods tend to treat a crowd either as a collection of individuals [21]–[23] or as an aggregated whole [8]–[10], [14].”
BackgroundRED: A Simple but Effective Baseline Predictor for the TrajNet Benchmark
Becker, S., Hug, R., Hübner, W., & Arens, M. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11015-4_13 · Cited by 70
“The BIWI Walking Pedestrians Dataset [36] also sometimes referenced as ETH Walking Pedestrians (EWAP), which is split into two sets (ETH and Hotel).”
BackgroundMethodHuman Trajectory Prediction using Spatially aware Deep Attention ModelsPreprint
Varshneya, D., & Srinivasaraghavan, G. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.09436 · Cited by 98
“We use three large scale multi-object tracking datasets - ETH [11] , UCY [12] and the Stanford Drone Dataset [13].”
MethodInfluentialEliminating Exposure Bias and Metric Mismatch in Multiple Object Tracking
Maksai, A., & Fua, P. (2019) · 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.2019.00477 · Cited by 69
“Many approaches rely on appearance [6, 17, 28, 57, 58, 12, 32, 46], motion [15], or social cues [20, 43].”
BackgroundSocial-aware Pedestrian Trajectory Prediction via States Refinement LSTM
Zhang, P., Xue, J., Zhang, P., et al. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3038217 · Cited by 69
Multi-Stream Representation Learning for Pedestrian Trajectory Prediction
Wu, Y., Wang, L., Zhou, S., Duan, J., Hua, G., & Tang, W. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i3.25389 · Cited by 68
“The average performance of five sub-sets in ETH-UCY is shown in Table 3.”
MethodInfluentialTracking by prediction: A deep generative model for multi-person localisation and tracking
Fernando, T., Denman, S., Sridharan, S., et al. (2018) · QUT ePrints (Queensland University of Technology) · Cited by 68
CSCNet: Contextual semantic consistency network for trajectory prediction in crowded spaces
Xia, B., Wong, C., Peng, Q., Yuan, W., & You, X. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108552 · Cited by 69
“Furthermore, it can be regarded as an essential component to recognize and analyze agents’ behaviors [11, 12].”
Social Interpretable Tree for Pedestrian Trajectory Prediction
Shi, L., Wang, L., Long, C., et al. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i2.20121 · Cited by 67
Trajectory-Prediction Techniques for Unmanned Aerial Vehicles (UAVs): A Comprehensive Survey
Shukla, P., Shukla, S., & Singh, A. K. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3471671 · Cited by 67
Multi-Camera Multi-Target Tracking with Space-Time-View Hyper-graph
Wen, L., Lei, Z., Chang, M., Qi, H., & Lyu, S. (2016) · International Journal of Computer Vision · doi:10.1007/s11263-016-0943-0 · Cited by 66
“As presented in Table 4, our approach achieves competitive performance with the state-of-the-art single-camera 3D multi-target tracking meth-ods (Leal-Taixé et al. 2011; Milan et al. 2015; Pellegrini et al. 2009; Klinger et al. 2015) according to the AvgRank score.”
MethodProvably safe motion of mobile robots in human environments
Liu, S. B., Roehm, H., Heinzemann, C., et al. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8202313 · Cited by 66
“3) using ground truth pedestrian trajectories from a labeled video source of a street scene in Zurich, Switzerland [14].”
MethodGD-GAN: Generative Adversarial Networks for Trajectory Prediction and Group Detection in Crowds
Fernando, T., Denman, S., Sridharan, S., & Fookes, C. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-20887-5_20 · Cited by 67
“The training, testing and validation splits for BIWI, CBE and VEIIG are taken from [26], [31] and [32] respectively.”
BackgroundMethodInfluentialSpatio-temporal multi-graph transformer network for joint prediction of multiple vessel trajectories
Liu, R. W., Zheng, W., & Liang, M.-H. (2024) · Engineering applications of artificial intelligence · doi:10.1016/j.engappai.2023.107625 · Cited by 69
Tracking as a Whole: Multi-Target Tracking by Modeling Group Behavior With Sequential Detection
Yuan, Y., Lu, Y., & Wang, Q. (2017) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2017.2686871 · Cited by 66
“Our approach is different than [13], [14] in that we model group behavior based on traffic force.”
MethodLearning and Inferring “Dark Matter” and Predicting Human Intents and Trajectories in Videos
Xie, D., Shu, T., Todorović, S., & Zhu, S. C. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2728788 · Cited by 70
“People’s trajectories have also been estimated based on inferring social interactions [27], [28], [29], [30], or detecting”
Analyzing the Variety Loss in the Context of Probabilistic Trajectory Prediction
Thiede, L., & Brahma, P. P. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.01005 · Cited by 65
“They, and many other recently published works [37, 36, 25, 28, 40], used the same MoN loss as a metric to benchmark their model against others, arguing that it is better suited to measure performance on multimodal data as compared to the widely used average displacement error [34].”
BackgroundSUMMIT: A Simulator for Urban Driving in Massive Mixed Traffic
Cai, P., Lee, Y., Luo, Y., & Hsu, D. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9197228 · Cited by 65
“Social force [15], [16], [17], [18] assume that traffic-agents are driven by attractive forces exerted by the destination and repulsive forces exerted by obstacles.”
BackgroundLearning Pedestrian Group Representations for Multi-modal Trajectory Prediction
Bae, I., Park, J.-H., & Jeon, H. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20047-2_16 · Cited by 65
Chen, X., Zhang, H., Hu, Y., et al. (2023) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2023.3273230 · Cited by 65
“ETH and UCY data: The two datasets [53], [54] consist of four traffic scenarios, five sets named Eth, Hotel, Univ, Zara1, and Zara 2.”
MethodGreedy Batch-Based Minimum-Cost Flows for Tracking Multiple Objects
Wang, X., Fan, B., Chang, S., Wang, Z., Liu, X., Tao, D., & Huang, T. S. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2723239 · Cited by 64
“For example, some approaches model a group of pedestrians by taking into account the psychical priors, such as the fact that they tend to form groups or they avoid to bump into each other [60]–[63].”
Human trajectory prediction in crowded scene using social-affinity Long Short-Term Memory
Pei, Z., Qi, X., Zhang, Y., Ma, M., & Yang, Y. (2019) · Pattern Recognition · doi:10.1016/j.patcog.2019.04.025 · Cited by 71
Which Way Are You Going? Imitative Decision Learning for Path Forecasting in Dynamic Scenes
Li, Y. (2019) · 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.2019.00038 · Cited by 63
“Namely Stanford Aerial Pedestrian (SAP) dataset [35], ETH dataset and UCY dataset [32] are selected.”
BackgroundReciprocal Learning Networks for Human Trajectory Prediction
Sun, H., Zhao, Z., & He, Z. (2020) · 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/cvpr42600.2020.00744 · Cited by 63
“We use the same error metrics in [1, 26] for performance evaluations.”
MethodMulti-target tracking by learning local-to-global trajectory models
Zhang, S., Wang, J., Wang, Z., Gong, Y., & Liu, Y. (2014) · Pattern Recognition · doi:10.1016/j.patcog.2014.08.013 · Cited by 65
“The tracking problem is solved by the data association [1–8] which strives to establish a unique identity for each target, and to simultaneously estimate the motion patterns of all targets and the assignment of detections to targets, by linking similar detections across frames.”
MethodA review of motion analysis methods for human Nonverbal Communication Computing
Metaxas, D., & Zhang, S. (2013) · Image and Vision Computing · doi:10.1016/j.imavis.2013.03.005 · Cited by 65
“[173] proposed a Linear Trajectory Avoidance (LTA) method to track multiple”
MethodCai, J., Zhang, Y., Yang, L., Cai, H., & Li, S. (2020) · Advanced Engineering Informatics · doi:10.1016/j.aei.2020.101173 · Cited by 66
A Set of Control Points Conditioned Pedestrian Trajectory Prediction
Bae, I., & Jeon, H.-G. (2023) · AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i5.25759 · Cited by 81
The Large-Scale Crowd Behavior Perception Based on Spatio-Temporal Viscous Fluid Field
Su, H., Yang, H., Zheng, S., et al. (2013) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2013.2277773 · Cited by 62
Social Attribute-Aware Force Model: Exploiting Richness of Interaction for Abnormal Crowd Detection
Zhang, Y., Qin, L., Ji, R., Yao, H., & Huang, Q. (2014) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2014.2355711 · Cited by 62
“modeling [12], people are driven by their own destination and emotion which is influenced by interactions with other people.”
BackgroundMethodInfluentialAn Evaluation of Trajectory Prediction Approaches and Notes on the TrajNet BenchmarkPreprint
Becker, S., Hug, R., Hübner, W., & Arens, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.07663 · Cited by 75
Detecting conversational groups in images and sequences: A robust game-theoretic approach
Vascon, S., Zemene, E., Cristani, M., et al. (2015) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2015.09.012 · Cited by 61
Dynamic and Static Context-Aware LSTM for Multi-agent Motion Prediction
Tao, C., Jiang, Q., Duan, L., & Luo, P. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58589-1_33 · Cited by 63
“The model with long queue length (q = 4) is the most suitable for the ETH dataset, where most of the trajectories are highly nonlinear.”
MethodInfluentialVehicle trajectory prediction and generation using LSTM models and GANs
Rossi, L., Ajmar, A., Paolanti, M., & Pierdicca, R. (2021) · PLoS ONE · doi:10.1371/journal.pone.0253868 · Cited by 66
IA-LSTM: Interaction-Aware LSTM for Pedestrian Trajectory Prediction
Yang, J., Chen, Y., Du, S., Chen, B., & Príncipe, J. C. (2024) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2024.3359237 · Cited by 70
“With tracking multiple people in complex scenarios, a powerful dynamic model [3] is still competitive in some datasets nowadays [1].”
BackgroundMethodInfluentialAction Anticipation by Predicting Future Dynamic Images
Rodríguez, C., Fernando, B., & Li, H. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11015-4_10 · Cited by 59
“other related problems in other domains of computer vision. Some of the work focuses on predicting (or forecasting) the future trajectories of pedestrians [20] or predicting motion from still images [20,21]. However, we are the rst to show the eectiveness of predicting good motion representations for early action anti…”
BackgroundAbnormal crowd behavior detection based on social attribute-aware force model
Zhang, Y., Qin, L., Yao, H., et al. (2012) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2012.6467453 · Cited by 60
“Inspired by the social behavior modeling [4], people are driven by their own destination and emotion which is influenced by the interaction of other people.”
MethodLong-Short Term Spatio-Temporal Aggregation for Trajectory Prediction
Yang, C., & Pei, Z. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3234962 · Cited by 60
Dynamic-group-aware networks for multi-agent trajectory prediction with relational reasoning
Xu, C., Wei, Y., Tang, B., et al. (2023) · Neural Networks · doi:10.1016/j.neunet.2023.11.005 · Cited by 60
Chandra, R., Bhattacharya, U., Roncal, C., Bera, A., & Manocha, D. (2019) · ACM Computer Science in Cars Symposium (CSCS) · doi:10.1145/3359999.3360495 · Cited by 59
“coordinates that form the future trajectory sequence. Examples of methods that explicitly model road-agent interaction include techniques based on social forces [18, 35], velocity obstacles [30], LTA [27], etc. Many of these models were designed to account for interactions between pedestrians in a crowd (i.e. homogene…”
MethodWolinski, D., Lin, M. C., & Pettré, J. (2016) · ACM Transactions on Graphics · doi:10.1145/2980179.2982442 · Cited by 59
“Later, in an algorithm derived from Social-Forces, [Karamouzas et al. 2009] used this point of closest approach as a source of repulsive forces; and [Pellegrini et al. 2009] used the distance of the closest approach to refrain from choosing velocities which might lead to collisions.”
BackgroundMethodvan der Aa, N. P., Luo, X., Giezeman, G.-J., Tan, R. T., & Veltkamp, R. C. (2011) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2011.6130396 · Cited by 59
“Examples can be found in multi-person tracking [14], and estimating poses from still images [6] and videos [11, 13].”
BackgroundLearning intentions for improved human motion prediction
Elfring, J., van de Molengraft, R., & Steinbuch, M. (2014) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2014.01.003 · Cited by 59
“Kitani et al. [11] ✓ × × ✓ Chen et al. [12] ✓ × × ✓ Pellegrini et al. [13] [14] × ✓ × ✓ Chung et al. [15] × ✓ × ✓ Luber et al. [16] × ✓ × ✓ Ziebart et al. [17] ✓”
Recurrent Network Models for Human DynamicsPreprint
Fragkiadaki, K., Levine, S., Felsen, P., & Malik, J. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1508.00271 · Cited by 89
“Similarly, for Computer Vision algorithms, predicting human motion is important for timely human-computer interaction [17], obstacle avoidance [22], and people tracking [8].”
BackgroundComplementary Attention Gated Network for Pedestrian Trajectory Prediction
Duan, J., Wang, L., Long, C., Zhou, S., Zheng, F., Shi, L., & Hua, G. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i1.19933 · Cited by 59
“Extensive experimental results on the ETH (Pellegrini et al. 2009) and UCY(Lerner, Chrysanthou, and Lischin-ski 2007) datasets show that our method outperforms all the competing state-of-the-art methods.”
MethodInfluentialRecurrent Encoder–Decoder Networks for Vessel Trajectory Prediction With Uncertainty Estimation
Capobianco, S., Forti, N., Millefiori, L. M., et al. (2022) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2022.3216823 · Cited by 58
“, over all the predicted positions of a trajectory) [37], [38].”
BackgroundVisual crowd surveillance through a hydrodynamics lens
Moore, B. E., Ali, S., Mehran, R., et al. (2011) · Communications of the ACM · doi:10.1145/2043174.2043192 · Cited by 58
“More recently, Pellegrini et al.18 expanded a social-force model to take into account destinations and desired directions of individuals, making it suitable for tracking individuals in crowded scenes.”
BackgroundRehman, A. U., Naqvi, S. M., Mihaylova, L. S., et al. (2015) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2015.2504340 · Cited by 58
“Therefore, the focus of this paper is to present a technique to robustly handle these challenges.”
MethodUsing road topology to improve cyclist path prediction
Pool, E. A. I., Kooij, J. F. P., & Gavrila, D. M. (2017) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2017.7995734 · Cited by 58
“Social factors, such as proximity to other VRUs, can be an informative factor [17], [18], [19].”
BackgroundA mobile robot that understands pedestrian spatial behaviors
Chung, S.-Y., & Huang, H. (2010) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2010.5649718 · Cited by 57
“Most developed methods [6] [13] only consider the reactive social forces which generate the next action of the pedestrian based on current observations.”
BackgroundUnderstanding Human Avoidance Behavior: Interaction-Aware Decision Making Based on Game Theory
Turnwald, A., Althoff, D., Wollherr, D., & BUSS, M. J. (2016) · International Journal of Social Robotics · doi:10.1007/s12369-016-0342-2 · Cited by 57
“2 For a detailed calculation please refer to [60].”
BackgroundHuman-Like Motion Planning Based on Game Theoretic Decision Making
Turnwald, A., & Wollherr, D. (2018) · International Journal of Social Robotics · doi:10.1007/s12369-018-0487-2 · Cited by 57
“The scene is a reproduction of the environment in the BIWI Walking Pedes-trians dataset [62] (Fig.”
BackgroundMethodTemporal Pyramid Network With Spatial-Temporal Attention for Pedestrian Trajectory Prediction
Li, Y., Liang, R., Wei, W., Wang, W. W., Zhou, J., & Li, X. (2021) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3065019 · Cited by 58
Interactive Model Predictive Control for Robot Navigation in Dense Crowds
Chen, Y., Zhao, F., & Lou, Y. (2021) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2020.3048964 · Cited by 57
SRAI-LSTM: A Social Relation Attention-based Interaction-aware LSTM for human trajectory prediction
Peng, Y., Zhang, G., Shi, J., et al. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.11.089 · Cited by 56
Set-Based Prediction of Pedestrians in Urban Environments Considering Formalized Traffic Rules
Koschi, M., Pek, C., Beikirch, M., et al. (2018) · International Conference on Intelligent Transportation Systems · doi:10.1109/itsc.2018.8569434 · Cited by 56
“Furthermore, we also validated our model using ground truth trajectories of 389 pedestrians from the publicly available BIWI Walking Pedestrians dataset of a street scene in Zurich, Switzerland [31].”
BackgroundMethodPedestrian Trajectory Prediction in Extremely Crowded Scenarios
Shi, X., Shao, X., Guo, Z., Wu, G., Zhang, H., & Shibasaki, R. (2019) · Sensors · doi:10.3390/s19051223 · Cited by 57
Dynamic Channel: A Planning Framework for Crowd Navigation
Cao, C., Trautman, P., & Iba, S. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794192 · Cited by 57
“For benchmarking, in our simulation we replay the human-trajectories from two public datasets: ETH [41] and UCY [42].”
MethodLi, J., Ma, H., & Tomizuka, M. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8793661 · Cited by 55
“Parameters a, b, c and d were generated randomly from U[3, 5] and the initial states were generated randomly from U[1, 3].”
MethodTemporal Pyramid Network for Pedestrian Trajectory Prediction with Multi-Supervision
Liang, R., Li, Y., Li, X., Tang, Y., Zhou, J., & Zou, W. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i3.16299 · Cited by 55
“Datasets: We evaluate our method on two public datasets, i.e., ETH(Pellegrini et al. 2009) and UCY(Lerner, Chrysanthou, and Lischinski 2007).”
BackgroundMethodDisentangled Multi-Relational Graph Convolutional Network for Pedestrian Trajectory Prediction
Bae, I., & Jeon, H.-G. (2021) · AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i2.16174 · Cited by 70
“In this section, we present our experimental results for two public datasets: ETH (Pellegrini et al. 2009) and UCY (Lerner, Chrysanthou, and Lischinski 2007) which contain pedestrian trajectories and various real-world human interactions.”
BackgroundResultFast Inference and Update of Probabilistic Density Estimation on Trajectory Prediction
Maeda, T., & Ukita, N. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00898 · Cited by 62
“Starting from deterministic approaches [26, 49], to tackle its indeterministic nature, many stochastic approaches are proposed, such as Gaussian distribution-based [1,42,66], generative adversarial networks (GANs)-based [9, 10, 25, 57], variational au-toencoder (VAE)-based [58, 71, 73],…”
BackgroundRealtime Multilevel Crowd Tracking Using Reciprocal Velocity Obstacles
Bera, A., & Manocha, D. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.714 · Cited by 55
“Other motion models used in pedestrian tracking are the Social Force model [28] [29] [3], LTA [30], RVO [31] and ATTR [32].”
MethodState of the Art Report on Video‐Based Graphics and Video Visualization
Borgo, R., Chen, M., Daubney, B., Grundy, E., Heidemann, G., Höferlin, B., Höferlin, M., Leitte, H., Weiskopf, D., & Xie, X. (2012) · Computer Graphics Forum · doi:10.1111/j.1467-8659.2012.03158.x · Cited by 77
Segmentation Based Particle Filtering for Real-Time 2D Object Tracking
Belagiannis, V., Schubert, F., Navab, N., & Ilić, S. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33765-9_60 · Cited by 57
“The test video sequences come from the following datasets: ETH Walking Pedestrians (EWAP) [26], Pedestrian dataset [27], Comets project [28] and the Aerial Action Dataset [29].”
MethodA Social Force Based Pedestrian Motion Model Considering Multi-Pedestrian Interaction with a Vehicle
Yang, D., Özgüner, Ü., & Redmill, K. (2020) · ACM Transactions on Spatial Algorithms and Systems · doi:10.1145/3373646 · Cited by 55
“To more accurately describe the effect of surrounding pedestrians, realistic pedestrian trajectory data [17, 25] was utilized to calibrate the parameters of the social force model.”
MethodCrowd Tracking with Dynamic Evolution of Group Structures
Zhu, F., Wang, X., & Yu, N.-H. (2014) · European Conference on Computer Vision · doi:10.1007/978-3-319-10599-4_10 · Cited by 54
“Recently, the social force model [9] has been used to track multiple pedestrians by modeling their interactions, the influence of destinations and scene structures with a physical model [19, 24].”
BackgroundCan Language Beat Numerical Regression? Language-Based Multimodal Trajectory Prediction
Bae, I., Lee, J., & Jeon, H.-G. (2024) · Computer Vision and Pattern Recognition · doi:10.1109/cvpr52733.2024.00078 · Cited by 93
“We conduct experiments on four public datasets: ETH [81], UCY [48], Stanford Drone Dataset (SDD) [93], and the Grand Central Station (GCS) [138] dataset to compare our LMTraj model with state-of-the-art trajectory predictors.”
MethodLi, K., Xu, Y., Wang, J., et al. (2019) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio49542.2019.8961764 · Cited by 54
“, constant velocity model [8], discrete choice model [9][10], social force model and its variants [11][12]) and data-driven methods (e.”
MethodHaghani, M. (2021) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2021.126145 · Cited by 55
Visual object tracking—classical and contemporary approaches
Ali, A., Jalil, A., Niu, J., Zhao, X., Rathore, S., Ahmed, J., & Iftikhar, M. A. (2015) · Frontiers of Computer Science · doi:10.1007/s11704-015-4246-3 · Cited by 56
“[195] use social force model and propose linear trajectory avoidance to predict motion patterns of individuals in crowd.”
MethodA Game-Theoretic Probabilistic Approach for Detecting Conversational Groups
Vascon, S., Zemene, E., Cristani, M., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16814-2_43 · Cited by 53
“Some of these principles informed early socially-motivated methods of people tracking [31] by the social force model [32], that originated from pedestrian simulation research.”
BackgroundMulti-agent path topology in support of socially competent navigation planning
Mavrogiannis, C., & Knepper, R. A. (2018) · The International Journal of Robotics Research · doi:10.1177/0278364918781016 · Cited by 53
MOANA: An Online Learned Adaptive Appearance Model for Robust Multiple Object Tracking in 3D
Tang, Z., & Hwang, J. (2019) · IEEE Access · doi:10.1109/access.2019.2903121 · Cited by 53
Vision-Based Trajectory Planning via Imitation Learning for Autonomous Vehicles
Cai, P., Sun, Y., Chen, Y., & Liu, M. (2019) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2019.8917149 · Cited by 52
“) is the final displacement error [26], meaning the distance between the planned and true final destination of a trajectory.”
BackgroundProceedings of the 24th Computer Vision Winter WorkshopPreprint
Haddad, S., Wu, M., Wei, H., et al. (2019) · arXiv (Cornell University) · doi:10.3217/978-3-85125-652-9 · Cited by 53
“Our evaluation is based on two widely-used datasets, ETH Walking Pedestrians (EWAP) [28], UCY Students and Zara [29].”
MethodA recurrent attention and interaction model for pedestrian trajectory prediction
Li, X., Liu, Y., Wang, K., et al. (2020) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2020.1003300 · Cited by 52
“Some works consider adding additional auxiliary information to help with trajectory prediction, such as the multi-task learning model named [12], scene-LSTM [13] and trajectory prediction with attribute and environment (TAE) [14].”
BackgroundOn Exposing the Challenging Long Tail in Future Prediction of Traffic Actors
Makansi, O., Çiçek, Ö., Marrakchi, Y., & Brox, T. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01290 · Cited by 56
“We trained the Trajectron++ (EWTA) with batch size 256 for 100 epochs in every EWTA stage ( k = K, ..., 1 ) for ETH-UCY and for 5 epochs in every EWTA stage for nuScenes.”
InfluentialWong, C., Xia, B., Zou, Z., et al. (2024) · 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/cvpr52733.2024.01798 · Cited by 52
“For example, the Social-Force-based methods [11, 20, 25, 32, 55] model and simulate agents’ behaviors mainly according to the rules in Newtonian mechanics.”
BackgroundMethodInfluentialRatsamee, P., Mae, Y., Ohara, K., et al. (2013) · International Journal of Humanoid Robotics · doi:10.1142/s0219843613500084 · Cited by 52
Social Force Model-Based MCMC-OCSVM Particle PHD Filter for Multiple Human Tracking
Feng, P., Wang, W., Dlay, S. S., Naqvi, S. M., & Chambers, J. A. (2016) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2016.2638206 · Cited by 52
“The distance dk (n) can be calculated as [16] dk (n) = ‖pk + tv k − pk − tv k ‖ (17)”
BackgroundMethodInfluentialMulti-camera tracking using a Multi-Goal Social Force Model
Mazzon, R., & Cavallaro, A. (2012) · Neurocomputing · doi:10.1016/j.neucom.2011.09.038 · Cited by 51
“A different microscopic approach for single-camera tracking is presented in [23] where Pellegrini et al. define the Linear Trajectory Avoidance (LTA) method.”
Method
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
