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 501–600 of 1,000
Wrong turn - No dead end: A stochastic pedestrian motion model
Pellegrini, S., Ess, A., Tanasković, M., & Gool, L. V. (2010) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2010.5543166 · Cited by 25
“While in [9] we show that taking into account these social factors improves prediction and object tracking performance compared to a constant velocity model, there are two main problems with the formulation: firstly, deterministically choosing the minimum of the energy function cannot account for the multiple options…”
BackgroundMethodResultInfluentialSelf-citationSelf-learning Trajectory Prediction with Recurrent Neural Networks at Intelligent Intersections
Bock, J., Beemelmanns, T., Klösges, M., & Kotte, J. (2017) · International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS) · doi:10.5220/0006374003460351 · Cited by 25
“In the machine learning research community, static publicly available datasets are used to compare the performance of different net architectures and to develop new models (Pellegrini et al., 2009; Robicquet et al.).”
MethodGibbs, J. A., Burgess, A. J., Pound, M. P., Pridmore, T., & Murchie, E. H. (2019) · PLANT PHYSIOLOGY · doi:10.1104/pp.19.00141 · Cited by 25
“Previous applications of MOT include the movement of people (i.e. pedestrians or sports players), vehicles, animals, or cells (Betke et al., 2000; Spampinato et al., 2008, 2012; Meijering et al., 2009; Pellegrini et al., 2009; Yang et al., 2011; Lu et al., 2013).”
MethodSimultaneous Identification and Tracking of Multiple People Using Video and IMUs
Henschel, R. D., von Marcard, T., & Rosenhahn, B. (2019) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00106 · Cited by 27
“Several works have focused on obtaining these weights from motion models [28, 33, 47, 8, 46, 27].”
BackgroundMulti-agent Trajectory Prediction with Fuzzy Query AttentionPreprint
Kamra, N., Zhu, H., Trivedi, D., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.15891 · Cited by 25
“Early work on predicting trajectories of multiple interacting agents dates back to more than two decades starting from Helbing and Molnar’s social force model [10] and its later extensions [19, 28] aimed at modeling behavior of humans in crowds, pedestrians on highways and vehicles on highways and freeways.”
BackgroundRecent trends in crowd management using deep learning techniques: a systematic literature review
Alasmari, A. M., Farooqi, N. S., & Alotaibi, Y. A. (2024) · Journal of Umm Al-Qura University for Engineering and Architecture · doi:10.1007/s43995-024-00071-3 · Cited by 24
"Seeing is Believing": Pedestrian Trajectory Forecasting Using Visual Frustum of Attention
Hasan, I., Setti, F., Tsesmelis, T., et al. (2018) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2018.00134 · Cited by 24
“We compare our method with four state-of-the-art model-based approaches, namely Linear Trajectory Avoidance (LTA) [29], Social Force model (SF) [45], Iterative Gaussian Process (IGP) [40], and multi-class Social Force model (SF-mc) [35].”
BackgroundMethodInfluentialSchaefer, S., Leung, K., Ivanovic, B., et al. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561443 · Cited by 24
CARPe Posterum: A Convolutional Approach for Real-Time Pedestrian Path Prediction
Mendieta, M., & Tabkhi, H. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i3.16335
“We evaluate our model on two widely used datasets in the path prediction domain, ETH (Pellegrini et al. 2009b) and UCY (Lerner, Chrysanthou, and Lischinski 2007).”
BackgroundMethodInfluentialFrom HRI to CRI: Crowd Robot Interaction—Understanding the Effect of Robots on Crowd Motion
Zhang, B., Amirian, J., Eberle, H., Pettré, J., Holloway, C., & Carlson, T. (2021) · International Journal of Social Robotics · doi:10.1007/s12369-021-00812-7 · Cited by 25
“Their solution was evaluated on the ETH pedestrian dataset [27].”
BackgroundMethodPreTR: Spatio-Temporal Non-Autoregressive Trajectory Prediction Transformer
Achaji, L., Barry, T., Fouqueray, T., Moreau, J., Aioun, F., & Charpillet, F. (2022) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc55140.2022.9922451 · Cited by 27
“We trained and evaluated the PreTR model on the ETH/UCY [31], [32] benchmark datasets for pedestrian multi-agent trajectory prediction.”
MethodWeng, X., Ivanovic, B., Kitani, K., & Pavone, M. (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/cvpr52688.2022.00646 · Cited by 28
“, [15, 40], and challenges (nuScenes [7], ETH/UCY [22, 27]), to enable a fair comparison.”
BackgroundMixed Traffic Trajectory Prediction Using LSTM–Based Models in Shared Space
Cheng, H., & Sester, M. (2018) · Lecture notes in geoinformation and cartography · doi:10.1007/978-3-319-78208-9_16 · Cited by 24
Fernando, T., Denman, S., Sridharan, S., & Fookes, C. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00149 · Cited by 25
“Following [11,20,21] we utilise the Stanford Drone (SD) dataset in our evaluation as it provides substantially lengthier trajectories compared to the frequently used Grand Central Station [25], and ETH-BIWI Walking Pedestrians [17] datasets.”
MethodDecentralized Multi-Agent Navigation Planning with Braids
Mavrogiannis, C., & Knepper, R. A. (2020) · Springer proceedings in advanced robotics · doi:10.1007/978-3-030-43089-4_56 · Cited by 24
“The social force model [9] and its variants [12, 17, 21] introduced a physics-inspired way of modeling pedestrian interactions: pedestrians are attracted to their destination and repulsed by obstacles or other agents.”
BackgroundBisagno, N., Saltori, C., Zhang, B., et al. (2020) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2020.103126 · Cited by 24
Near-Online Multi-target Tracking with Aggregated Local Flow DescriptorPreprint
Choi, W. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1504.02340 · Cited by 24
“(e.g. A1 = {d1, d10, d23}) during the time of presence; i.e. (V T1 ,DT1 )→ A.”
Beyond particle flow: Bag of Trajectory Graphs for dense crowd event recognition
Zhang, Y., Qin, L., Yao, H., et al. (2013) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2013.6738737 · Cited by 23
Multiple player tracking in basketball court videos
Fu, X., Zhang, K., Wang, C., & Fan, C. (2020) · Journal of Real-Time Image Processing · doi:10.1007/s11554-020-00968-x · Cited by 24
“In crowded scenarios, pedestrian motion will be more complex, and researchers adopt the more expressive Social Force Model [24, 31, 38, 46].”
BackgroundMulti-Camera Trajectory Forecasting: Pedestrian Trajectory Prediction in a Network of Cameras
Styles, O., Guha, T., Sánchez, V., et al. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00516 · Cited by 23
“Existing datasets commonly used for trajectory forecasting, such as ETH [8] and UCY [6], consist of just a single camera view and are therefore unsuitable for MCTF.”
MethodTrajectory Prediction Network for Future Anticipation of Ships
Dijt, P., & Mettes, P. (2020) · International Conference on Multimedia Retrieval · doi:10.1145/3372278.3390676 · Cited by 23
“Social Ways uses GANs to generate pedestrian trajectories and has state-of-the-art performance in the trajectory prediction dataset BIWI Walking Pedestrians [30].”
BackgroundModeling social interaction and intention for pedestrian trajectory prediction
Chen, K., Song, X., & Ren, X. (2021) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2021.125790 · Cited by 23
SCAN: A Spatial Context Attentive Network for Joint Multi-Agent Intent Prediction
Sekhon, J., & Fleming, C. H. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i7.16762 · Cited by 23
“These datasets contain annotated trajectories of pedestrians interacting in several social situations and include challenging behavior such as collision avoidance, movement in groups, yielding right of way, couples walking together, groups crossing groups, etc.(Pellegrini et al. 2009).”
BackgroundMethodInfluentialAn efficient Spatial–Temporal model based on gated linear units for trajectory prediction
Liu, S., Wang, Y., Sun, J., et al. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.12.051 · Cited by 23
Learning Sparse Interaction Graphs of Partially Detected Pedestrians for Trajectory Prediction
Huang, Z., Li, R., Shin, K., & Driggs-Campbell, K. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3138547 · Cited by 37
“We use two publicly available trajectory datasets for benchmarking: ETH [31] and UCY [32].”
MethodMulti-Agent Trajectory Prediction with Graph Attention Isomorphism Neural Network
Liu, Y., Qi, X., Sisbot, E. A., et al. (2022) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv51971.2022.9827155 · Cited by 23
Bimodal Extended Kalman Filter-Based Pedestrian Trajectory Prediction
Lin, C.-Y., Kau, L., & Chan, C. (2022) · Sensors · doi:10.3390/s22218231 · Cited by 23
Multi-view Tracking Using Weakly Supervised Human Motion Prediction
Engilberge, M., Liu, W., & Fua, P. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00163 · Cited by 23
“Modeling human motion as flow when tracking people has been a concern long before the advent of deep learning [47, 57, 11, 37, 14, 34, 20, 10, 45, 42, 3, 9].”
BackgroundZhu, D., Khan, Q., & Cremers, D. (2024) · Neurocomputing · doi:10.1016/j.neucom.2023.127220 · Cited by 28
Using maximum consistency context for multiple target association in wide area traffic scenes
Shi, X., Li, P., Ling, H., et al. (2013) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2013.6638042 · Cited by 23
“To address such ambiguity, researchers have investigated ways to integrate additional information, such as prior knowledge [1], scene structures [6], and target context [7, 10, 11].”
BackgroundA Survey on Video-based Graphics and Video Visualization
Borgo, R., Chen, M., Daubney, B., Grundy, E., Heidemann, G., Höferlin, B., Höferlin, M., Jänicke, H., Weiskopf, D., & Xie, X. (2011) · doi:10.2312/eg2011/stars/001-023 · Cited by 23
“For the domain of tracking individuals in crowded environments, models of social interaction have been learnt to predict how people will behave which can be used to improve the performance of tracking algorithms [PESvG09].”
MethodRobot Navigation Based on Human Trajectory Prediction and Multiple Travel Modes
Chen, Z., Song, C., Yang, Y., Zhao, B., Hu, Y., Liu, S., & Zhang, J. (2018) · Applied Sciences · doi:10.3390/app8112205 · Cited by 23
“In this paper, we compare our proposed improved socially conscious model with the state-of-the-art Social-LSTM in [8] with the datasets, ETH [24] and UCY [25].”
BackgroundMethodCrossing-Road Pedestrian Trajectory Prediction via Encoder-Decoder LSTM
Xue, P., Liu, J., Chen, S., Zhou, Z., Huo, Y., & Zheng, N. (2019) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2019.8917510 · Cited by 22
“So, the results predicted by our method shows less error compared with the actual path, and the values of ADE and FDE are significantly lower than those of the other two methods.”
BackgroundMethodInfluentialSocNav1: A Dataset to Benchmark and Learn Social Navigation Conventions
Manso, L. J., Núñez, P., Calderita, L. V., et al. (2020) · Aston Publications Explorer (Aston University) · Cited by 23
MCENET: Multi-Context Encoder Network for Homogeneous Agent Trajectory Prediction in Mixed Traffic
Cheng, H., Liao, W., Yang, M. Y., Sester, M., & Rosenhahn, B. (2020) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc45102.2020.9294296 · Cited by 24
“Besides the validation in mixed traffic, we also validate our method on pedestrian benchmark datasets ETH [27] and UCY [24].”
MethodInfluentialReciprocal Twin Networks for Pedestrian Motion Learning and Future Path Prediction
Sun, H., Zhao, Z., Yin, Z., & He, Z. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3076078 · Cited by 25
Forecasting Events Using an Augmented Hidden Conditional Random Field
Wei, X., Lucey, P., Vidas, S., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16817-3_37 · Cited by 23
Measuring Collectiveness via Refined Topological Similarity
Li, X., Chen, M., & Wang, Q. (2016) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/2854000 · Cited by 22
“Pellegrini et al. [2009] introduced a dynamic social behavior model to track multi-individuals from a vehicle-mounted camera.”
BackgroundKong, L., Huang, D., & Wang, Y. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3009034 · Cited by 23
“According to whether future information is required to process the current frame, they are divided into two groups, i.e. online matching based [14]–[19] and offline association based [20]– [24].”
BackgroundHow Can I See My Future? FvTraj: Using First-Person View for Pedestrian Trajectory Prediction
Bi, H., Zhang, R., Mao, T., Deng, Z., & Wang, Z. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58571-6_34 · Cited by 24
“In terms of our proposed module-based architecture, we found that incorporating both the social-aware attention module and the view-aware attention module can significantly improve the performances on ETH, HOTEL, and UNIV.”
BackgroundMethodInfluentialA Two-Block RNN-Based Trajectory Prediction From Incomplete Trajectory
Fujii, R., Vongkulbhisal, J., Hachiuma, R., & Saitô, H. (2021) · IEEE Access · doi:10.1109/access.2021.3072135 · Cited by 25
Dynamic pedestrian trajectory forecasting with LSTM-based Delaunay triangulation
Ma, Q., Zou, Q., Huang, Y., & Wang, N. (2021) · Applied Intelligence · doi:10.1007/s10489-021-02562-5 · Cited by 22
Propagating State Uncertainty Through Trajectory Forecasting
Ivanovic, B., Lin, Y., Shrivastava, S., Chakravarty, P., & Pavone, M. (2022) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra46639.2022.9811776 · Cited by 24
“The ETH, UCY, and nuScenes datasets evaluate our approach’s ability to model real-world pedestrians and vehicles.”
BackgroundMethodInfluentialD-STGCN: Dynamic Pedestrian Trajectory Prediction Using Spatio-Temporal Graph Convolutional Networks
Sighencea, B. I., Stanciu, R., & Căleanu, C. D. (2023) · Electronics · doi:10.3390/electronics12030611 · Cited by 22
WOMD-LiDAR: Raw Sensor Dataset Benchmark for Motion Forecasting
Chen, K., Ge, R., Qiu, H., Ai-Rfou, R., Qi, C. R., Zhou, X., Yang, Z., Ettinger, S. M., Sun, P., Leng, Z., et al. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610651 · Cited by 22
“There has been an increasing number of motion forecasting datasets released [17], [26], [25], [9], [47], [49], [39], [15], [36], [30], [4], [8], [6].”
BackgroundUsing the Agoraset dataset: Assessing for the quality of crowd video analysis methods
Courty, N., Allain, P., Creusot, C., & Corpetti, T. (2014) · Pattern Recognition Letters · doi:10.1016/j.patrec.2014.01.004 · Cited by 23
Predicting Ball Ownership in Basketball from a Monocular View Using Only Player Trajectories
Wei, X., Sha, L., Lucey, P., et al. (2015) · doi:10.1109/iccvw.2015.106 · Cited by 22
“In terms of tracking and predicting behaviors of multiple agents, an abundance of work has recently focussed on the topic due to the influx of real-world data sources and a myriad of useful applications, most notably in the crowd and security domains [1, 2, 3, 4, 5, 6, 7].”
BackgroundProbabilistic multi-person localisation and tracking in image sequences
Klinger, T., Rottensteiner, F., & Heipke, C. (2017) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2016.11.006 · Cited by 22
Pedestrian Group Detection in Shared Space
Cheng, H., Li, Y., & Sester, M. (2019) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2019.8813849 · Cited by 22
“…0.91 0.92 0.91 0.91 0.92 0.91 0.9 0.89 0.81 0.47 0.85 0.85 0.87 0.89 0.89 0.9 0.92 0.91 0.92 0.92 0.91 0.9 0.9 0.83 0.47 0.82 0.81 0.83 0.84 0.87 0.87 0.88 0.88 0.89 0.9 0.9 0.92 0.93 0.83 0.47 We run the time–sequence DBSCAN cluster on the BIWI benchmark datasets [27] to further test its utility.”
InfluentialMF-SORT: Simple Online and Realtime Tracking with Motion Features
Fu, H., Wu, L., Jian, M., Yang, Y., & Wang, X. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-34120-6_13 · Cited by 22
Accurate Long-Term Multiple People Tracking Using Video and Body-Worn IMUs
Henschel, R. D., von Marcard, T., & Rosenhahn, B. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3013801 · Cited by 23
“Several works have focused on obtaining these weights from motion models [11], [33], [37]–[40].”
BackgroundTracking Beyond Detection: Learning a Global Response Map for End-to-End Multi-Object Tracking
Wan, X., Cao, J., Zhou, S., Wang, J., & Zheng, N. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3113169 · Cited by 23
End-to-End Pedestrian Trajectory Forecasting with Transformer Network
Yao, H., Wan, W., & Li, X. (2022) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi11010044 · Cited by 22
“The detail information of the datasets and metrics are listed as follows: • Datasets; Following the related prior research, we evaluate the proposed method on two public datasets: ETH [40] and UCY [41].”
MethodPedestrian trajectory prediction using goal-driven and dynamics-based deep learning framework
Wang, H., Zhi, W., Batista, G., & Chandra, R. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.126557 · Cited by 22
Labeled multi-bernoulli track-before-detect for multi-target tracking in video
Rathnayake, T., Gostar, A. K., Hoseinnezhad, R., et al. (2015) · RMIT Research Repository (RMIT University Library) · Cited by 21
$L_0$ Regularized Stationary-Time Estimation for Crowd Analysis
Yi, S., Wang, X., Lu, C., et al. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2560807 · Cited by 21
“Existing works focus on detecting motion patterns of crowds [1], [3], [4], [5] and analyzing the interactions between moving pedestrians [6], [7], [8], [9].”
BackgroundAutomatic Topic Discovery for Multi-Object Tracking
Luo, W., Stenger, B., Zhao, X., & Kim, T. (2015) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v29i1.9789 · Cited by 22
“There has been significant progress in multi-object tracking (Zhang, Li, and Nevatia 2008; Pellegrini et al. 2009; Xing, Ai, and Lao 2009; Pirsiavash, Ramanan, and Fowlkes 2011; Milan, Schindler, and Roth 2013; Luo et al. 2014; Leal-Taix ´ e et al. 2014), however, issues like tracking management,…”
BackgroundMethodInfluentialContinual Pedestrian Trajectory Learning With Social Generative Replay
Wu, Y., Bighashdel, A., Chen, G., Dubbelman, G., & Jancura, P. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3231833 · Cited by 21
“The proposed benchmark consists of four tasks ( N = 4 ), each of which corresponds to a pedestrian trajectory prediction problem on a specific dataset: 1) ETH [25], 2) UCY [26], 3) inD [27], and 4) INTERACTION [28].”
MethodLearning a Group-Aware Policy for Robot Navigation
Katyal, K. D., Gao, Y., Markowitz, J., Pohland, S., Rivera, C. G., Wang, I.-J., & Huang, C. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9981183 · Cited by 24
“[7], [21], [22], [23], [24], [25] and used handcrafted rules as planning constraints [8], [26], [27], [28].”
TraGCAN: Trajectory Prediction of Heterogeneous Traffic Agents in IoV Systems
Li, J., Shi, H., Guo, Y., et al. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3228818 · Cited by 21
“We considered the following common metrics to evaluate the performance of different traffic agent trajectory prediction algorithms [25], [38], [41].”
MethodEnd-To-End Compression for Surveillance Video With Unsupervised Foreground-Background Separation
Zhao, Y. Q., Luo, D., Wang, F., Gao, H., Ye, M., & Zhu, C. (2023) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2023.3280039 · Cited by 21
“…follows the work in [36], all experiments are conducted on the surveillance video CUHK Square Dataset [55] with one hour and EWAP Dataset [56] with twenty minutes, where the EWAP Dataset contains two surveillance video datasets, EWAP_Eth and EWAP_Hotel, including pedestrians, vehicles…”
MethodSTM-GCN: a spatiotemporal multi-graph convolutional network for pedestrian trajectory prediction
Taki, Y., Zemmouri, E. M., & Bouzid, A. (2023) · The Journal of Supercomputing · doi:10.1007/s11227-023-05467-x · Cited by 24
“Experiments with the ETH [7] and UCY [8] benchmarking datasets show that STM-GCN significantly improves pedestrian trajectory prediction, with 6% in terms of the average displacement error (ADE) and 10% in terms of the final displacement error (FDE).”
MethodInfluentialBio-inspired relevant interaction modelling in cognitive crowd management
Chiappino, S., Morerio, P., Marcenaro, L., et al. (2014) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-014-0224-0 · Cited by 21
“As shown for instance in (Mehran et al. 2009; Pellegrini et al. 2009; Luber et al. 2010; Mazzon et al. 2013), a social force model describing interactions among the individual members of a group of people has been proposed to detect abnormal events in crowd videos.”
BackgroundLong-Term Activity Forecasting Using First-Person Vision
Bokhari, S. Z., & Kitani, K. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-54193-8_22 · Cited by 21
“In particular, there has been significant work concentrated on predicting very short term actions, on the order of a few seconds [1] [2] [3] [4].”
BackgroundMultiperson Tracking by Online Learned Grouping Model With Nonlinear Motion Context
Chen, X., Qin, Z., An, L., & Bhanu, B. (2015) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2015.2511480 · Cited by 21
Li, H., Luo, B., Song, W., & Yang, C. (2023) · Neural Networks · doi:10.1016/j.neunet.2023.06.007 · Cited by 21
Pedestrian Travel Time Estimation in Crowded Scenes
Yi, S., Li, H., & Wang, X. (2015) · IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.359 · Cited by 20
“The social force model [10] has been used for pedestrian simulation [9], pedestrian tracking [18], pedestrian interaction analysis [21], and abnormal behavior detection [16].”
MethodDensePeds: Pedestrian Tracking in Dense Crowds Using Front-RVO and Sparse Features
Chandra, R., Bhattacharya, U., Bera, A., & Manocha, D. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8968470 · Cited by 20
“Other motion models used in pedestrian tracking are the Social Force model [28], LTA [29], and ATTR [30].”
MethodChoi, D., Min, K.-W., & Choi, J. (2020) · IET Computer Vision · doi:10.1049/iet-cvi.2019.0546 · Cited by 20
IAN: Multi-Behavior Navigation Planning for Robots in Real, Crowded Environments
Dugas, D., Nieto, J. I., Siegwart, R., et al. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341457 · Cited by 20
“For example, Pfeif-fer et al. [13] train a long short-term memory (LSTM) neural network on several pedestrian datasets [14] to predict human trajectories.”
BackgroundPishgu: Universal Path Prediction Network Architecture for Real-time Cyber-physical Edge Systems
Noghre, G. A., Katariya, V., Pazho, A. D., et al. (2023) · ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS) · doi:10.1145/3576841.3585933 · Cited by 20
“Two widely used pedestrian path prediction datasets are ETH [50] and UCY [34].”
BackgroundMethodResultInfluentialCSIR: Cascaded Sliding CVAEs With Iterative Socially-Aware Rethinking for Trajectory Prediction
Zhou, H., Yang, X., Ren, D., Huang, H. Y., & Fan, M. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3300730 · Cited by 21
“The ETH/UCY [20], [21] dataset contains two subsets of five scenes (ETH: ETH and HOTEL; UCY: UNIV, ZARA1, and ZARA2) with 1536 unique pedestrians and thousands of nonlinear trajectories.”
MethodVulnerable road user detection and safety enhancement: A comprehensive survey
Silva, R. M., Azevedo, G. F., Berto, M. V. V., Rocha, J. R., Fidelis, E. C., Nogueira, M. V., Lisboa, P., & Almeida, T. A. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.128529 · Cited by 30
“Several studies have explored these approaches.”
Dual-channel CNN for efficient abnormal behavior identification through crowd feature engineering
Xu, Y., Lü, L., Xu, Z., et al. (2018) · Machine Vision and Applications · doi:10.1007/s00138-018-0971-6 · Cited by 20
Social-VRNN: One-Shot Multi-modal Trajectory Prediction for Interacting PedestriansPreprint
Brito, B., Zhu, H., Pan, W., & Alonso–Mora, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.09056 · Cited by 24
“Moreover, we evaluate our method on real data using the publicly available datasets [29, 19].”
Bridging Requirements, Planning, and Evaluation: A Review of Social Robot Navigation
Karwowski, J., Szynkiewicz, W., & Niewiadomska‐Szynkiewicz, E. (2024) · Sensors · doi:10.3390/s24092794 · Cited by 24
Point-based path prediction from polar histograms
Coscia, P., Castaldo, F., Palmieri, F., et al. (2016) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · Cited by 19
Social Graph Transformer Networks for Pedestrian Trajectory Prediction in Complex Social Scenarios
Liu, Y., Yao, L., Li, B., Wang, X., & Sammut, C. (2022) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3511808.3557455 · Cited by 21
“We perform ablation experiments on the ETH dataset.”
MethodSafe Robot Navigation in a Crowd Combining NMPC and Control Barrier Functions
Vulcano, V., Tarantos, S. G., Ferrari, P., et al. (2022) · Proceedings of the IEEE Conference on Decision & Control, including the Symposium on Adaptive Processes · doi:10.1109/cdc51059.2022.9993397 · Cited by 19
“Existing human prediction algorithms are based on either social-force (e.g., [20]) or data-driven models (e.g., [21]).”
BackgroundJoint Metrics Matter: A Better Standard for Trajectory Forecasting
Weng, E., Hoshino, H., Ramanan, D., et al. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01858 · Cited by 19
“We evaluate on the commonly-used pedestrian trajectory datasets ETH [38] / UCY [29], used across the field in nearly all pedestrian trajectory forecasting works.”
MethodIMGCN: interpretable masked graph convolution network for pedestrian trajectory prediction
Chen, W., Sang, H., Wang, J., et al. (2024) · Transportmetrica B Transport Dynamics · doi:10.1080/21680566.2024.2389896 · Cited by 19
“The ETH datasets encompass two scenarios: ETH and HOTEL.”
InfluentialTHÖR-MAGNI: A large-scale indoor motion capture recording of human movement and robot interaction
Schreiter, T., Almeida, T., Zhu, Y., et al. (2024) · The International Journal of Robotics Research · doi:10.1177/02783649241274794 · Cited by 19
“Bosch GmbH, Corporate Research, Stuttgart, Germany { andrey.rudenko, luigi.palmieri, kaioliver.arras } @de.bosch.com 3 Mobile Robotics Group, Department of Electrical Engineering and Automation, Aalto University, Finland tomasz.kucner@aalto.fi In this work, we follow on and further develop the TH ¨…”
Unsupervised Trajectory Modelling Using Temporal Information via Minimal Paths
Cancela, B., Iglesias, A., Ortega, M., 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.2014.327 · Cited by 19
Pedestrian Trajectory Prediction Based on Transfer Learning for Human-Following Mobile Robots
Akabane, R., & Kato, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3111917 · Cited by 23
“These methods perform short-term predictions of trajectories for tracking targets, for example, using gait velocity and generating an extended Kalman fil-ter (KF) by integrating the proximity effect of mutual interaction between persons with the motion model [9].”
MethodPTPGC: Pedestrian trajectory prediction by graph attention network with ConvLSTM
Yang, J., Sun, X., Wang, R. G., & Xue, L. X. (2021) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2021.103931 · Cited by 20
Efficient Multiple People Tracking Using Minimum Cost Arborescences
Henschel, R. D., Leal-Taixé, L., & Rosenhahn, B. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-11752-2_21 · Cited by 21
“[19] take avoidance behavior of humans into account, and – Leal-Taixé et al.”
BackgroundPedestrian Path Forecasting in Crowd
Li, Y. (2017) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3123266.3123287 · Cited by 19
“All these datasets involve very challenging crowd scenarios as suggested in [21].”
BackgroundMethodInfluentialLearning and predicting on-road pedestrian behavior around vehicles
Deo, N., & Trivedi, M. M. (2017) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2017.8317865 · Cited by 19
“Some approaches [9], [10], [12] termed ’social forces’ explicitly model humanhuman interactions such as collision avoidance, intra-group attraction, as well as attraction to goals.”
BackgroundRecurrent Network Models for Kinematic Tracking.Preprint
Fragkiadaki, K., Levine, S., & Malik, J. (2015) · arXiv (Cornell University) · Cited by 19
Ryu, K., & Mehr, N. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610404 · Cited by 18
Zhou, X., Chen, X., & Yang, J. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3509954 · Cited by 18
“• HEAT-I-R [52]: This paper extracts dynamics using type-specific encoders, processes inter-agent interactions via HEAT, and shares map features through a selective gate mechanism.”
Zhou, X., Chen, X., & Yang, J. J. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110125 · Cited by 19
Chen, W., Sang, H., & Zhao, Z. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3556839 · Cited by 18
“These three datasets include: ETH and UCY Datasets: The ETH [59] and UCY [60] datasets are the classical datasets for the pedestrian trajectory prediction task.”
A Non-parametric Hierarchical Model to Discover Behavior Dynamics from Tracks
Kooij, J. F. P., Englebienne, G., & Gavrila, D. M. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33783-3_20 · Cited by 18
“(Left column) Tracks (in random colors) and a screenshot from the BIWI dataset seq eth [20].”
BackgroundMethodInfluentialExtended social force model‐based mean shift for pedestrian tracking under obstacle avoidance
Zhang, X., Zhang, X., Wang, Y., et al. (2016) · IET Computer Vision · doi:10.1049/iet-cvi.2016.0022 · Cited by 18
“As for the public ETH Walking Pedestrians (EWAP) dataset [34, 35], the resolution of each image is 720 × 576 pixels.”
BackgroundBaG: Behavior-Aware Group Detection in Crowded Urban Spaces Using WiFi Probes
Shen, J., Cao, J., & Liu, X. (2020) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2020.2999491 · Cited by 18
Move Beyond Trajectories: Distribution Space Coupling for Crowd Navigation
Sun, M., Baldini, F., Trautman, P., et al. (2021) · doi:10.15607/rss.2021.xvii.053 · Cited by 18
“For evaluation, we considered the crowd datasets ETH [31] and UCY [26] and crowd simulators based on the the social forces model (SFM, [18]; e.g., PEDSIM [15]) and optimal reciprocal collision avoidance agents [47, 46], such as implemented in [4].”
MethodPUTS: Privacy-Preserving and Utility-Enhancing Framework for Trajectory Synthesization
Sun, X., Ye, Q., Hu, H., et al. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2023.3288154 · Cited by 18
Dimensionless numbers reveal distinct regimes in the structure and dynamics of pedestrian crowds
Cordes, J., Schadschneider, A., & Nicolas, A. (2024) · PNAS Nexus · doi:10.1093/pnasnexus/pgae120 · Cited by 21
“…(single-file motion (23), bottleneck flows (24), corridor flows (25 – 27), antipodal scenarios (28)) and empirical observations in outdoor settings (29, 30); further details about these scenarios and the way we have smoothed out head sways from the trajectories can be found in SI Appendix A.”
BackgroundFeng, A., Han, C., Gong, J., Yi, Y., Qiu, R., & Yang, C. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3421373 · Cited by 18
“Early trajectory prediction relies heavily on the third-person bird’s-eye view, which is characterized by a fixed scene and numerous agents [5], [6], [7].”
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
