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 301–400 of 1,000
Learning to Predict Human Behavior in Crowded Scenes
Alahi, A., Ramanathan, V., Goel, K., Robicquet, A., Sadeghian, A., Fei-Fei, L., & Savarese, S. (2017) · Elsevier eBooks · doi:10.1016/b978-0-12-809276-7.00011-4 · Cited by 57
“This is similar to the setting used by [10, 11].”
BackgroundMethodResultInfluentialPedestrian Trajectory Prediction Combining Probabilistic Reasoning and Sequence Learning
Li, Y., Lu, X., Wang, J., & Li, K. (2020) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2020.2966117 · Cited by 57
“Pellegrini et al. defined the expected point of closest approachtocomputetheenergypotentials[13].”
BackgroundRepresenting Multimodal Behaviors With Mean Location for Pedestrian Trajectory Prediction
Shi, L., Wang, L., Long, C., Zhou, S., Tang, W., Zheng, N., & Hua, G. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3268110 · Cited by 56
“ETH [13] and UCY [14] are the most widely used benchmarks for pedestrian trajectory prediction.”
BackgroundMethodInfluentialLearning How Pedestrians Navigate: A Deep Inverse Reinforcement Learning Approach
Fahad, M., Chen, Z., & Guo, Y. P. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593438 · Cited by 50
“• Average displacement error is the mean square error (MSE) between the simulated trajectory and the measured trajectory [14], [27].”
BackgroundMethodTRiPOD: Human Trajectory and Pose Dynamics Forecasting in the Wild
Adeli, V., Ehsanpour, M., Reid, I. D., Niebles, J. C., Savarese, S., Adeli, E., & Rezatofighi, H. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01314 · Cited by 49
Safety-Compliant Generative Adversarial Networks for Human Trajectory Forecasting
Kothari, P., & Alahi, A. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2022.3233906 · Cited by 51
“We evaluate the performance of our architecture against several state-of-theart methods on the ETH/UCY datasets [58], [59] and on the interaction-centric TrajNet++ benchmark [40].”
MethodWho knows who - Inverting the Social Force Model for finding groups
Sochman, J., & Hogg, D. (2011) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2011.6130338 · Cited by 49
“The values of the parameters Θ used in our implementation of the Social Force Model are listed in Table 1 together with references to the source of those particular values.”
MethodA Simultaneous Multi-Object Zooming System Using an Ultrafast Pan-Tilt Camera
Hu, S., Shimasaki, K., Jiang, M., Senoo, T., & Ishii, I. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3054425 · Cited by 50
“Multi-object detection and tracking [48]–[50] is an important methodology for localizing multiple objects in video sequences to be used in computer vision applications, such as face detection [51]–[53], pedestrian tracking [54]–[56], and vehicle traffic monitoring [57]–[59].”
MethodSMEMO: Social Memory for Trajectory Forecasting
Marchetti, F., Becattini, F., Seidenari, L., & Bimbo, A. D. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3356755 · Cited by 62
“…[18]; a representation of the environment should be obtained in order to provide spatial cues or constraints on admissible areas [19], [20], [21], [22], [23]; other agents must be taken into account to detect social patterns such as group dynamics [3], [4], [5], [6], [7], [8], [19], [24], [25].”
BackgroundChen, W., Sang, H., Wang, J., et al. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3525080 · Cited by 48
“We train and test the model on the ETH [53], UCY [54], and SDD [55] datasets.”
Modeling, Simulation and Visual Analysis of Crowds: A Multidisciplinary Perspective
Ali, S., Nishino, K., Manocha, D., & Shah, M. A. (2013) · The Kluwer international series in video computing · doi:10.1007/978-1-4614-8483-7_1 · Cited by 48
A real time algorithm for people tracking using contextual reasoning
Lascio, R. D., Foggia, P., Percannella, G., et al. (2013) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2013.04.004 · Cited by 48
Automatic detection and tracking of pedestrians from a moving stereo rig
Konrad Schindler, B., Ess, A., Leibe, B., & Gool, L. V. (2010) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2010.06.006 · Cited by 48
“…motion planning638 of real people is not independent of their environment, so including models of639 social behaviour such as those developed for crowd simulation (e.g. Helbing and640 Molnár, 1995; Schadschneider, 2001) could potentially improve dynamic models641 (c.f. Pellegrini et al., 2009).”
BackgroundSelf-citationUnderstanding Human Behaviors in Crowds by Imitating the Decision-Making Process
Zou, H., Su, H., Song, S., & Zhu, J. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12316 · Cited by 48
“Admittedly, inferring the aforementioned decision-making process of pedestrians is challenging since it is complex and determined by various factors, including personal intentions, neighboring pedestrians, and physical obstacles (Pellegrini et al. 2009).”
BackgroundMethodChen, X., Zhang, H., Deng, F., Liang, J. S., & Yang, J. J. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3342040 · Cited by 47
“To evaluate the performance of our proposed prediction model, we perform experiments on three well-established crowd scene datasets with rich social interactions, namely, ETH [57], UCY [58], and the Stanford Drone dataset (SDD) [59].”
MethodBinary Quadratic Programing for Online Tracking of Hundreds of People in Extremely Crowded Scenes
Dehghan, A., & Shah, M. A. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2687462 · Cited by 46
“Although the effect of neighbors in tracking was used before in [36], its extension to high-density crowds was explored first in [11].”
BackgroundGoal-driven Long-Term Trajectory Prediction
Tran, H., Le, V., & Tran, T. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00084 · Cited by 51
“These datasets contain real-world annotated trajectories in five different scenes: ETH, HOTEL (from ETH dataset), ZARA1, ZARA2, UNIV (from UCY dataset).”
MethodMultimodal interaction-aware motion prediction for autonomous street crossing
Radwan, N., Burgard, W., & Valada, A. (2020) · The International Journal of Robotics Research · doi:10.1177/0278364920961809 · Cited by 46
AMENet: Attentive Maps Encoder Network for trajectory prediction
Cheng, H., Liao, W., Yang, M. Y., Rosenhahn, B., & Sester, M. (2021) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2020.12.004 · Cited by 51
“The most popular pedestrian scenes ETH [46] and UCY [47] are also included in the benchmark.”
BackgroundStructured learning for detection of social groups in crowd
Solera, F., Calderara, S., & Cucchiara, R. (2013) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2013.6636608 · Cited by 45
“We tested our system on two publicly available datasets, namely the BIWI Walking Pedestrians dataset [16] and the Crowds-By-Examples (CBE) dataset [11].”
MethodPredicting Primary Gaze Behavior Using Social Saliency Fields
Park, H. S., Jain, E., & Sheikh, Y. A. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.435 · Cited by 44
Multi-Object Tracking Using Online Metric Learning with Long Short-Term Memory
Wan, X., Wang, J., Kong, Z., Zhao, Q. Q., & Shun-Ming, D. E. N. G. (2018) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2018.8451174 · Cited by 45
“Later on, lots of works have been combining various of components such as motion dynamics and interaction information to obtain an effective state estimation [11, 12, 13, 14, 15, 16, 17].”
BackgroundEvoSTGAT: Evolving spatiotemporal graph attention networks for pedestrian trajectory prediction
Tang, H., Wei, P., Li, J., & Zheng, N. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.03.051 · Cited by 47
Learning Behavior Patterns from Video: A Data-driven Framework for Agent-based Crowd Modeling
Zhong, J., Cai, W., Luo, L., et al. (2015) · Adaptive Agents and Multi-Agents Systems · doi:10.5555/2772879.2773256 · Cited by 44
ATPFL: Automatic Trajectory Prediction Model Design under Federated Learning Framework
Wang, C., Chen, X., Wang, J., & Wang, H. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.00645 · Cited by 45
“In order to evaluate the performance of our approach under FL framework, we construct a federated TP dataset D by equally distributing trajectory data of two publicly available datasets: ETH [23] and UCY [15], to 20 clients.”
MethodVision and Trajectory–Based Dynamic Collision Prewarning Mechanism for Tower Cranes
Zhang, M., & Ge, S. (2022) · Journal of Construction Engineering and Management · doi:10.1061/(asce)co.1943-7862.0002309 · Cited by 46
Crowd Management Through Optimal Layout of Fences: An Ant Colony Approach Based on Crowd Simulation
Liao, X.-C., Chen, W., Guo, X.-Q., Zhong, J., & Hu, X.-M. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3272318 · Cited by 43
“automatic crowd modeling [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35].”
Autonomous crowds tracking with box particle filtering and convolution particle filtering
Freitas, A. D., Mihaylova, L. S., Gning, A., et al. (2016) · Automatica · doi:10.1016/j.automatica.2016.03.009 · Cited by 43
New Multiple-Target Tracking Strategy Using Domain Knowledge and Optimization
Ding, R., Yu, M., Oh, H., et al. (2016) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2016.2615188 · Cited by 43
“Examples of microscopic models include car-following model [28], cellular automata model [29], optimization-based models [30], and force-based models [12].”
MethodPedestrian trajectory prediction via the Social‐Grid LSTM model
Cheng, B., Xu, X., Zeng, Y., Ren, J., & Jung, S. (2018) · The Journal of Engineering · doi:10.1049/joe.2018.8316 · Cited by 43
Forecasting People Trajectories and Head Poses by Jointly Reasoning on Tracklets and Vislets
Hasan, I., Setti, F., Tsesmelis, T., Belagiannis, V., Amin, S., Bue, A. D., Cristani, M., & Galasso, F. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2949414 · Cited by 49
“Metric Dataset SF [10] LTA [2] Vanilla”
BackgroundMethodInfluentialRole playing learning for socially concomitant mobile robot navigation
Li, M., Jiang, R., Ge, S. S., & Lee, T. H. (2018) · CAAI Transactions on Intelligence Technology · doi:10.1049/trit.2018.0008 · Cited by 40
“In [11], linear trajectory avoidance is developed as a dynamic model for pedestrians in video space for short-term trajectory prediction and it is integrated into visual tracking system.”
BackgroundMethodInfluentialMultiple Object Forecasting: Predicting Future Object Locations in Diverse Environments
Styles, O., Guha, T., & Sánchez, V. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093446 · Cited by 43
“Several datasets have been created explicitly for pedestrian trajectory forecasting, such as UCY [23], ETH [29], and Stanford Drone [35].”
MethodKhurana, T., Dave, A., & Ramanan, D. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00316 · Cited by 43
“For STGAT, the 8- and 12-timestep models used are trained on the ETH [39] dataset and for SGAN, the 8- and 12-timestep models are trained on the ZARA1 [29] dataset.”
BackgroundMethodExploring Dynamic Context for Multi-path Trajectory Prediction
Cheng, H., Liao, W., Tang, X., Yang, M. Y., Sester, M., & Rosenhahn, B. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9562034 · Cited by 43
“Trajnet is a superset of diverse popular benchmark datasets: ETH [48], UCY [49], Stanford Drone Dataset [50], BIWI Hotel [48], and MOT PETS [51].”
BackgroundSpatial-Temporal-Spectral LSTM: A Transferable Model for Pedestrian Trajectory Prediction
Zhang, C., Ni, Z., & Berger, C. (2023) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2023.3285804 · Cited by 48
“…positions in the real world are pre-processed into 2-dimensional ( x, y ) positions from the bird’s-eye-view, and used as ground truth for training and evaluation. b) ETH and UCY datasets ETH [50] and UCY [51] datasets are widely used by existing studies on pedestrian trajectory prediction.”
MethodSocial Self-Attention Generative Adversarial Networks for Human Trajectory Prediction
Yang, C., Pan, H., Sun, W., & Gao, H. (2023) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2023.3299899 · Cited by 42
“There are a variety of datasets in the field of autonomous driving,suchasETH[21],UCY[22],andTrajNet[47].”
BackgroundMethodL0 Regularized Stationary Time Estimation for Crowd Group Analysis
Yi, S., Wang, X., Lu, C., & Jia, J. (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.2014.284 · Cited by 42
“Existing work focuses on detecting motion patterns of crowds [25, 1, 26, 18, 32, 13, 30, 31, 20] and analyzing interactions among pedestrians during movement [15, 17, 19, 28, 20].”
BackgroundVelocity obstacle based local collision avoidance for a holonomic elliptic robot
Lee, B. H., Jeon, J. D., & Oh, J. H. (2016) · Autonomous Robots · doi:10.1007/s10514-016-9580-2 · Cited by 42
Reactive Planning for Assistive Robots
Bevilacqua, P., Frego, M., Fontanelli, D., et al. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2795642 · Cited by 42
“Our reactive-planning has been validated in two ways: a set of software simulations, where the dynamic obstacles trajectories were captured from a real scenario recorded at ETH [33] and the simulation emulates the sensor acquisition from the walker to generate the clothoid model discussed in Section II; with a direct…”
MethodSSeg-LSTM: Semantic Scene Segmentation for Trajectory Prediction
Syed, A., & Morris, B. (2019) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2019.8813801 · Cited by 43
A Location-Velocity-Temporal Attention LSTM Model for Pedestrian Trajectory Prediction
Xue, H., Huynh, D. Q., & Reynolds, M. A. (2020) · IEEE Access · doi:10.1109/access.2020.2977747 · Cited by 42
“The combined ETH [6] and UCY [53] dataset is another widely used public dataset for evaluating trajectory prediction methods.”
MethodInfluentialForeground-Background Parallel Compression With Residual Encoding for Surveillance Video
Wu, L., Huang, K., Shen, H., & Gao, L. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3027741 · Cited by 42
“E XPERIMENTS A. Experiment Setup 1) Dataset: We train our model on surveillance video clips taken from CUHK Square Dataset [34] and EWAP Dataset [35].”
Bialkowski, A., Lucey, P., Carr, P., et al. (2014) · Cited by 42
Particle-based Pedestrian Path Prediction using LSTM-MDL Models
Hug, R., Becker, S., Hübner, W., & Arens, M. (2018) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2018.8569478 · Cited by 43
“This excludes popular datasets like ETH [18] or UCY [19].”
BackgroundHuang, L., Zhuang, J., Cheng, X., Xu, R., & Ma, H. (2021) · IEEE Access · doi:10.1109/access.2021.3069134 · Cited by 42
“On the public real datasets UCY and ETH, the maximum number of pedestrians per frame is 65, and the model can still accurately predict the future trajectory.”
MethodInfluentialLv, K., & Yuan, L. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3283544 · Cited by 42
“The ETH dataset includes two scenarios (ETH and HOTEL), whereas the UCY dataset includes three scenarios (ZARA-01, ZARA-02, and UCY).”
BackgroundMethodInfluentialMFAN: Mixing Feature Attention Network for trajectory prediction
Li, J., Yang, L., Chen, Y., et al. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.109997 · Cited by 41
Socially competent navigation planning by deep learning of multi-agent path topologies
Mavrogiannis, C., Blukis, V., & Knepper, R. A. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8206601 · Cited by 41
Under the hood of transformer networks for trajectory forecasting
Franco, L., Placidi, L., Giuliari, F., Hasan, I., Cristani, M., & Galasso, F. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.109372 · Cited by 40
An Online Learned Elementary Grouping Model for Multi-target Tracking
Chen, X., Qin, Z., An, L., et al. (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.2014.162 · Cited by 40
“In [19] a more effective dynamic model leveraging nearby people’s positions is proposed.”
BackgroundTracking in dense crowds using prominence and neighborhood motion concurrence
Idrees, H., Warner, N., & Shah, M. A. (2013) · Image and Vision Computing · doi:10.1016/j.imavis.2013.10.006 · Cited by 40
Joint Prediction for Kinematic Trajectories in Vehicle-Pedestrian-Mixed Scenes
Bi, H., Fang, Z., Mao, T., et al. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.01048 · Cited by 40
“Inspired by the previous work [25], we chose ADE (Average Displacement Error: the average Euclidean distance error between the predicted result and ground truth over the whole sequence) and FDE (Final Displacement Error: the Euclidean distance error at the last step between the predicted result and ground truth) as th…”
BackgroundMethodLong-Term Pedestrian Trajectory Prediction Using Mutable Intention Filter and Warp LSTM
Huang, Z., Hasan, A., Shin, K., Li, R., & Driggs-Campbell, K. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.3047731 · Cited by 40
“time step of the prediction period [47].”
DROGON: A Trajectory Prediction Model based on Intention-Conditioned Behavior ReasoningPreprint
Choi, C., Malla, S., Patil, A., & Choi, J. H. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.00024 · Cited by 40
A review on crowd analysis of evacuation and abnormality detection based on machine learning systems
Bahamid, A., & Ibrahim, A. M. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-07758-5 · Cited by 40
Wang, R., Song, X., Hu, Z., et al. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3229733 · Cited by 39
“The proposed model is evaluated on two public ETH [44] and UCY [45] datasets.”
MethodMSN: Multi-Style Network for Trajectory Prediction
Wong, C., Xia, B., Peng, Q., et al. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3274777 · Cited by 39
“This phenomenon occurs more in the ETH-UCY dataset but less in the SDD, which also verifies our point of view.”
BackgroundMethodInfluentialForecasting Pedestrian Trajectory with Machine-Annotated Training Data
Styles, O., Ross, A. A., & Sánchez, V. (2019) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2019.8814207 · Cited by 40
“In particular, deep learning models have been applied to trajectory forecasting in a surveillance setting with a fixed overhead camera on datasets such as UCY [18] and ETH [22], or forecasting vehicle trajectories [34].”
MethodForecasting Time-to-Collision from Monocular Video: Feasibility, Dataset, and Challenges
Manglik, A., Weng, X., Ohn-Bar, E., & Kitani, K. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8967730 · Cited by 44
“While most of the existing datasets for human trajectory prediction are from a fixed overhead camera [11], [12], our dataset of 13,658 video segments targets the first-person view which is more intuitive for mobile robots.”
BackgroundXu, Y., Ren, D., Li, M., Chen, Y., Fan, M., & Xia, H. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3057326 · Cited by 40
“In this section, we demonstrate the experimental results on two public pedestrian-walking datasets: ETH [41] and UCY [42].”
Nishimura, H., Ivanovic, B., Gaidon, A., et al. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341469 · Cited by 39
“For brevity, we refer to these as the “ETH,” “HOTEL,” and “UNIV” sequences, respectively.”
MethodInfluentialAn Adaptive Motion Model for Person Tracking with Instantaneous Head-Pose Features
Baxter, R. H., Leach, M., Mukherjee, S. S., et al. (2014) · IEEE Signal Processing Letters · doi:10.1109/lsp.2014.2364458 · Cited by 39
“Others have incorporated the idea that socially grouped pedestrians will attempt to stay in close proximity [10], [11].”
BackgroundRadar-Assisted Predictive Beamforming for UAV-Aided Networks: A Deep-Learning Solution
Yin, R., Peng, J., Cai, Y., Wu, C., Champagne, B., & Al‐Dhahir, N. (2025) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2025.3572037 · Cited by 44
“All experiments are performed on the publicly available pedestrian trajectory: ETH [39] dataset.”
Sidestepping Intractable Inference with Structured Ensemble Cascades
Weiss, D. A., Sapp, B., & Taskar, B. (2010) · ScholarlyCommons (University of Pennsylvania) · Cited by 38
SteerFit: automated parameter fitting for steering algorithms
Berseth, G., Kapadia, M., Haworth, B., et al. (2015) · Symposium on Computer Animation · doi:10.5555/2849517.2849536 · Cited by 38
Driver and pedestrian awareness-based collision risk analysis
Roth, M., Flohr, F. B., & Gavrila, D. M. (2016) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2016.7535425 · Cited by 38
“The related Linear Trajectory Avoidance model [17] for short-term path prediction uses the expected point of closest approach to foreshadow and avoid possible collisions.”
BackgroundHuman Re-identification in Crowd Videos Using Personal, Social and Environmental Constraints
Assari, S. M., Idrees, H., & Shah, M. A. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-46475-6_8 · Cited by 38
“Social Force Models have been used for improving tracking performance [25, 38,52].”
BackgroundMethodBruijning, M., Visser, M. D., Hallmann, C. A., & Jongejans, E. (2018) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.12975 · Cited by 41
Location-Velocity Attention for Pedestrian Trajectory Prediction
Xue, H., Huynh, D. Q., & Reynolds, M. A. (2019) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2019.00221 · Cited by 38
“It contains various trajectory-based activity forecasting datasets such as BIWI [33], UCY [26], MOT [13], and Stanford Drone Dataset (SDD) [35].”
BackgroundMethodMulti-Object Tracking and Segmentation Via Neural Message Passing
Brasó, G., Cetintas, O., & Leal-Taixé, L. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01678-6 · Cited by 39
“Data association can be done on a frame-by-frame basis for online applications (Breitenstein et al., 2009; Ess et al., 2008; Pellegrini et al., 2009) or track-by-track (Berclaz et al.”
Human Trajectory Prediction with Momentary Observation
Sun, J., Li, Y., Chai, L., Fang, H.-S., Li, Y., & Lu, C. (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/cvpr52688.2022.00636 · Cited by 39
“Exhaustive experiments are conducted on ETH/UCY dataset [13,27] and SDD [28] (Stanford Drone Dataset).”
MethodAdaPT: Real-time adaptive pedestrian tracking for crowded scenes
Bera, A., Galoppo, N., Sharlet, D., Lake, A., & Manocha, D. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907095 · Cited by 39
A New Hybrid Synthetic Aperture Imaging Model for Tracking and Seeing People Through Occlusion
Yang, T., Zhang, Y., Tong, X., Zhang, X., & Yu, R. (2013) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2013.2242553 · Cited by 38
Deciphering the Crowd: Modeling and Identification of Pedestrian Group Motion
Yücel, Z., Zanlungo, F., Ikeda, T., et al. (2013) · Sensors · doi:10.3390/s130100875 · Cited by 38
“BIWI Walking Pedestrians dataset contains two sequences, BIWI-ETH and BIWI-Hotel, recorded from birds-eye view with a total of 650 tracks over 20 minutes [20].”
MethodZhou, L., Zhao, Y., Yang, D., et al. (2022) · Applied Intelligence · doi:10.1007/s10489-021-02997-w · Cited by 37
“are not walking alone [19], and their interactions are not just one-on-one.”
BackgroundDynamic Attention-Based CVAE-GAN for Pedestrian Trajectory Prediction
Zhou, Z., Huang, G., Su, Z., et al. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3231531 · Cited by 37
“In this section, we evaluate the proposed DACG on two widely used benchmarks for pedestrian trajectory prediction, i.e., ETH [37] and UCY [38], which cover rich movement patterns and human-human interactions.”
MethodSTGlow: A Flow-Based Generative Framework With Dual-Graphormer for Pedestrian Trajectory Prediction
Liang, R., Li, Y., Zhou, J., & Li, X. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3294998 · Cited by 44
Sparse Instance Conditioned Multimodal Trajectory Prediction
Dong, Y., Wang, L., Zhou, S., & Hua, G. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00895 · Cited by 37
“We evaluate our proposed method on two benchmark datasets, i.e. , ETH-UCY [31, 21] and Stanford Drone Dataset (SDD) [33].”
MethodBi-label Propagation for Generic Multiple Object Tracking
Luo, W., Kim, T., Stenger, B., et al. (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.2014.168 · Cited by 37
“Inspired by crowd simulation models, a dynamic model considering social motion patterns is introduced in [21].”
MethodStructural Constraint Data Association for Online Multi-object Tracking
Yoon, J. H., Lee, C., Yang, M., & Yoon, K. (2018) · International Journal of Computer Vision · doi:10.1007/s11263-018-1087-1 · Cited by 38
“Trackers: KalmanSFM (Pellegrini et al. 2009), LPSFM (Leal-Taixé et al.”
BackgroundTrajectory Prediction by Coupling Scene-LSTM with Human Movement LSTM
Huynh, M., & Alaghband, G. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-33720-9_19 · Cited by 37
“ene Information LSTM Network. 1 Introduction Human movement trajectory prediction is an essential task in computer vision with applications in autonomous driving cars [11], robotic navigation systems [14,19], and intelligent human tracking systems [9,12]. Given the past movement trajectories of pedestrians in a video…”
BackgroundMethodInfluentialAC-VRNN: Attentive Conditional-VRNN for multi-future trajectory prediction
Bertugli, A., Calderara, S., Coscia, P., et al. (2021) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2021.103245 · Cited by 37
“We investigate the ability of Vanilla VRNNs to predict accurate trajectories on ETH, UCY and SDD datasets.”
BackgroundMethodInfluentialOpenTraj: Assessing Prediction Complexity in Human Trajectories Datasets
Amirian, J., Zhang, B., Castro, F. V., et al. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-69544-6_34 · Cited by 37
“es of computer vision, mobile robotics, and crowd dynamics have been noticeably active on this topic. Many outstanding prediction algorithms have been proposed, from physics-based social force models [1,2,3] to data-driven models [4,5,6]. In parallel, eorts have been made towards a proper benchmarking of the existing…”
InfluentialKolekar, S., Gite, S., Pradhan, B., & Kotecha, K. V. (2021) · IEEE Access · doi:10.1109/access.2021.3116303 · Cited by 37
“The ETH [112] and UCY [113] datasets are the most widely used literature predicting pedestrian trajectories.”
BackgroundWang, R., Hu, Z., Song, X., et al. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2023.3329676 · Cited by 36
“ETH and UCY: ETH [38] and UCY [39] are two different public benchmarks for validating the prediction effectiveness of pedestrian trajectories.”
BackgroundKaramouzas, I., Sohre, N. E., Hu, R., et al. (2018) · ACM Transactions on Graphics · doi:10.1145/3272127.3275079 · Cited by 36
“These include: i) sparse crowds (average space-time density less than 0.5 ppl/m 2 ) captured at a snowy intersection [Pellegrini et al. 2009] and in front of a church [Lerner et al. 2010], labeled Snow_eth and Church , respectively; ii) Medium-density (between 0.5-2 ppl/m 2 ) groups of shoppers…”
BackgroundInfluentialEuro-PVI: Pedestrian Vehicle Interactions in Dense Urban Centers
Bhattacharyya, A., Reino, D. O., Fritz, M., & Schiele, B. (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/cvpr46437.2021.00634 · Cited by 36
“Datasets like ETH/UCY [23] and Stanford Drone [32] are among the first datasets in the field of pedestrian trajectory prediction.”
BackgroundUnlimited Neighborhood Interaction for Heterogeneous Trajectory Prediction
Zheng, F., Wang, L., Zhou, S., Tang, W., Niu, Z., Zheng, N., & Hua, G. (2021) · IEEE International Conference on Computer Vision · doi:10.1109/iccv48922.2021.01292 · Cited by 41
Yao, R., Zeng, W., Chen, Y., & He, Z. (2021) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2021.103415 · Cited by 35
Multimodal Pedestrian Trajectory Prediction Using Probabilistic Proposal Network
Chen, W., Yang, Z.-G., Xue, L., Duan, J., Sun, H.-B., & Zheng, N. (2023) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2022.3229694 · Cited by 33
“Extensive experimental results show that: PPNet outperforms Y-net on ETH [22]/UCY [23] benchmark by 7 .”
MethodSocial navigation model based on human intention analysis using face orientation
Ratsamee, P., Mae, Y., Ohara, K., Kojima, M., & Arai, T. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696575 · Cited by 36
“Most research pays attention to how to integrate these factors via a number of different models [3], [4], [5], [6].”
BackgroundMonocular Long-Term Target Following on UAVs
Li, R., Pang, M., Zhao, C., Zhou, G., & Fang, L. (2016) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2016.11 · Cited by 36
“…tracking method so as to push consequential applications of UAVs far beyond amusement to surveillance [5], augmented reality [10], behavior modeling [23] etc. Contrary to the GPS-based target following on UAVs, which requires the target to wear an extra GPS-equipped device for communication [26]…”
BackgroundZhu, W., Liu, Y.-H., Wang, P., Zhang, M., Wang, T., & Yi, Y. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.109772 · Cited by 38
Expanding the Deployment Envelope of Behavior Prediction via Adaptive Meta-Learning
Ivanovic, B., Harrison, J., & Pavone, M. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10161155 · Cited by 35
“The ETH and UCY datasets are a standard benchmark in the field, comprised of pedestrian trajectories captured at 2.5 Hz ( t = 0.4s) in Zurich and Cyprus, respectively.”
MethodInfluentialYang, B., Yan, K., Hu, C., Hu, H., Yu, Z., & Ni, R. (2024) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2024.3487255 · Cited by 35
“We evaluate DSBCL-PRM on the ETH/UCY and SDD dataset [36], [37], [38], which is a widely used benchmark for pedestrian trajectory prediction.”
Vision-based motion prediction for construction workers safety in real-time multi-camera system
Jeon, Y., Tran, D. Q., Kulinan, A. S., Kim, T., Park, M., & Park, S. (2024) · Advanced Engineering Informatics · doi:10.1016/j.aei.2024.102898 · Cited by 38
Learning behavior patterns from video for agent-based crowd modeling and simulation
Zhong, J., Cai, W., Luo, L., et al. (2016) · Autonomous Agents and Multi-Agent Systems · doi:10.1007/s10458-016-9334-8 · Cited by 35
A Computationally Efficient Model for Pedestrian Motion Prediction
Batkovic, I., Zanon, M., Lübbe, N., & Falcone, P. (2018) · European Control Conference (ECC) · doi:10.23919/ecc.2018.8550300 · Cited by 35
“This enables predicting agent movements in crowded environments [11], [12], [13].”
BackgroundCF-LSTM: Cascaded Feature-Based Long Short-Term Networks for Predicting Pedestrian Trajectory
Xu, Y., Yang, J., & Du, S. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i07.6943 · Cited by 36
“…has many applications in various fields such as autonomous driving (Lerner, Chrysanthou, and Lischinski 2010), robot navigation (Luo et al. 2018; Pellegrini et al. 2009; Vivacqua et al. 2017; Thrun et al. 2002) and surveillance camera analysis (Leonard and Durrant-Whyte 1990; Trautman and…”
Background
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
