└── README.md /README.md: -------------------------------------------------------------------------------- 1 | ## Awesome-Trajectory-Prediction 2 | 3 | 4 | 5 | 本仓库由[公众号【自动驾驶之心】](https://mp.weixin.qq.com/s?__biz=Mzg2NzUxNTU1OA==&mid=2247542481&idx=1&sn=c6d8609491a128233c3c3b91d68d22a6&chksm=ceb80b18f9cf820e789efd75947633aec9d2f1e8b58c29e5051c05a64b21ae63c244d54886a1&token=11182364&lang=zh_CN#rd)团队整理，欢迎关注，一览最前沿的技术分享! 6 | 7 | 自动驾驶之心是国内首个自动驾驶开发者社区！这里有最全面有效的自动驾驶与AI学习路线（感知/定位/融合）和自动驾驶与AI公司内推机会！ 8 | 9 | ## Other awesomes 10 | 11 | [colorfulfuture/Awesome-Trajectory-Motion-Prediction-Papers (github.com)](https://github.com/colorfulfuture/Awesome-Trajectory-Motion-Prediction-Papers) 12 | 13 | ## 一、Pedestrian trajectory prediction correlation 14 | 15 | ### 1. Summary of pedestrian trajectory prediction methods 16 | 17 | Y-Net: From goals, waypoints & paths to long term human trajectory forecasting, ICCV 2021 18 | 19 | [[code]](https://github.com/HarshayuGirase/Human-Path-Prediction/tree/master/ynet) 20 | 21 | MG-GAN: A Multi-Generator Model Preventing Out-of-Distribution Samples in Pedestrian TrajectoryPrediction, ICCV 2021 22 | 23 | [[code]](https://github.com/selflein/MG-GAN) 24 | 25 | Semantic Synthesis of Pedestrian Locomotion, ACCV 2020 26 | 27 | Spatio-Temporal Graph Transformer Networks for Pedestrian Trajectory Prediction，ECCV2020 28 | 29 | [[code]](https://github.com/Majiker/STAR) 30 | 31 | SimAug: Learning Robust Representations from 3D Simulation for Pedestrian Trajectory Prediction inUnseen Cameras 32 | 33 | [[code]](https://github.com/JunweiLiang/Multiverse) 34 | 35 | Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human TrajectoryPrediction,2020 CVPR, 36 | 37 | [[code]](https://github.com/abduallahmohamed/Social-STGCNN/) 38 | 39 | Disentangling Human Dynamics for Pedestrian Locomotion Forecasting with Noisy Supervision 40 | 41 | PIE:ALarge-Scale Dataset and Models for Pedestrian Intention Estimation and Trajectory Prediction 42 | 43 | STGAT: Modeling Spatial-Temporal Interactions for Human Trajectory Prediction, 2019 ICCV 44 | 45 | Trajectory Prediction by Coupling Scene-LSTM with Human Movement LSTM, 2019 ISVC 46 | 47 | SEABIG:A Deep Learning-Based Method for Location Prediction in Pedestrian Semantic Trajectories 48 | 49 | A novel model based on deep learning for Pedestrian detection and Trajectory prediction 50 | 51 | Pedestrian Trajectory Prediction Using a Social Pyramid, 2019 PRICAI 52 | 53 | Human Trajectory Prediction using Adversarial Loss, 2019(from Alahi, conference unknown fornow 54 | 55 | Social Ways: Learning Multi-Modal Distributions of Pedestrian Trajectories with GANs 56 | 57 | [[code]](https://github.com/crowdbotp/socialways) 58 | 59 | Peeking into the Future: Predicting Future Person Activities and Locations in Videos 60 | 61 | [[code]](https://github.com/google/next-prediction) 62 | 63 | Future Person Localization in First-Person Videos, 2018 CVPR 64 | 65 | [[code]](https://github.com/takumayagi/fpl) 66 | 67 | Move, Attend and Predict: Anattention-based neural model for people's movement prediction, 2018 PatternRecognition Letters 68 | 69 | Group LSTM: Group Trajectory Prediction in Crowded Scenarios,2018 ECCV Workshop 70 | 71 | Pedestrian Trajectory Prediction in Extremely Crowded Scenarios, 2019 Sensors (journal) 72 | 73 | The Simpler the Better: Constant Velocity for Pedestrian Motion Prediction, 2019 74 | SR-LSTM:State Refinement for LSTM towards Pedestrian Trajectory Prediction, 2019 CVPR 75 | 76 | Situation-Aware Pedestrian Trajectory Prediction with Spatio-Temporal Attention Model, 2019 ComputerVision Winter Workshop (cvw) 77 | Depth Information Guided Crowd Counting for Complex Crowd Scenes, 2018 78 | 79 | GD-GAN: Generative Adversarial Networks for Trajectory Prediction and Group Detection in Crowds, 2018 ACCV 80 | 81 | Tracking by Prediction: A Deep Generative Model for Mutli-Person Localisation and Tracking, 2018 WACV 82 | 83 | “Seeing is Believing”: Pedestrian Trajectory Forecasting Using Visual Frustum of Attention, 2018 WACV 84 | 85 | Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks, 2018 CVPR 86 | 87 | [[code]](https://github.com/agrimgupta92/sgan) 88 | 89 | Long-Term On-Board Prediction of People in Traffic Scenes under Uncertainty, 2018 CVPR 90 | 91 | [[code]](https://github.com/apratimbhattacharyya18/onboard_long_term_prediction) 92 | 93 | Encoding Crowd Interaction with Deep Neural Network for Pedestrian Trajectory Prediction, 2018 CVPR 94 | 95 | [[code]](https://github.com/svip-lab/CIDNN) 96 | 97 | Scene-LSTM: A Model for Human Trajectory Prediction, 2018 ArXiv 98 | 99 | Bi-prediction: pedestrian trajectory prediction based on bidirectional LSTM classification, 2017 DICTA 100 | 101 | Human Trajectory Prediction using Spatially aware Deep Attention Models, 2017 arxiv 102 | 103 | Soft + Hardwired Attention: An LSTM Framework for Human Trajectory Prediction and Abnormal Event Detection, 2017 arxiv 104 | 105 | Forecasting Interactive Dynamics of Pedestrians with Fictitious Play, 2017 CVPR 106 | 107 | Social LSTM: Human Trajectory Prediction in Crowded Spaces, 2016 CVPR 108 | 109 | STF-RNN: Space Time Features-based Recurrent Neural Network for predicting People Next Location, 2016 SSCI 110 | 111 | [[code]](https://github.com/mhjabreel/STF-RNN) 112 | 113 | ## 二、Summary of pedestrian trajectory prediction methods 114 | 115 | ### 1. Vehicle trajectory prediction 116 | 117 | PTNet: Physically Feasible Vehicle Trajectory Prediction, NeurIPS ML4AD Workshop 118 | 119 | Injecting Knowledge in Data-driven Vehicle Trajectory Predictors, 2021 Transportation research part C 120 | 121 | [[code]](https://github.com/vita-epfl/RRB) 122 | 123 | Multiple Futures Prediction, 2019 NeurIPS 124 | 125 | Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs, 2019 arXiv 126 | 127 | [[code]](https://gamma.umd.edu/researchdirections/autonomousdriving/spectralcows/) 128 | 129 | RobustTP: End-to-End Trajectory Prediction for Heterogeneous Road-Agents in Dense Traffic with Noisy Sensor Inputs, 2019 ACM CSCS 130 | 131 | [[code]](https://github.com/rohanchandra30/TrackNPred) 132 | 133 | TraPHic: Trajectory Prediction in Dense and Heterogeneous Traffic Using Weighted Interactions, 2019 CVPR 134 | 135 | [[code]](https://github.com/rohanchandra30/TrackNPred) 136 | 137 | Multi-Step Prediction of Occupancy Grid Maps with Recurrent Neural Networks, 2019 CVPR 138 | 139 | Argoverse: 3D Tracking and Forecasting With Rich Maps, 2019 CVPR 140 | 141 | Robust Aleatoric Modeling for Future Vehicle Localization, 2019 CVPR 142 | 143 | Convolutional Social Pooling for Vehicle Trajectory Prediction 144 | 145 | [[code]](https://github.com/nachiket92/conv-social-pooling) 146 | 147 | Trajectory-Prediction with Vision: A Survey 148 | 149 | Spatio-Temporal Trajectory Similarity Measures: A Comprehensive Survey and Quantitative Study 150 | 151 | Multimodal Trajectory Prediction: A Survey 152 | 153 | 154 | 155 | ## 三、methodology 156 | 157 | ### 1. Summary of classical methods (including physics-based, machine learning, deep learning, reinforcement learning) 158 | 159 | #### Physics-based approach 160 | 161 | ##### Single track 162 | 163 | Vehicle dynamics and external disturbance estimation for vehicle path prediction 164 | 165 | Reducing navigation errors by planning with realistic vehicle model 166 | 167 | Situation assessment of an autonomous emergency brake for arbitrary vehicle-to-vehicle collision scenarios 168 | 169 | Cooperative path prediction in vehicular environments 170 | 171 | Model-based threat assessment for avoiding arbitrary vehicle collisions 172 | 173 | An adaptive peer-to-peer collision warning system 174 | 175 | A multilevel collision mitigation approach-its situation assessment, decision making, and performance tradeoffs 176 | 177 | ##### Kalman filtering 178 | 179 | Real time trajectory prediction for collision risk estimation between vehicles 180 | 181 | Recognition of dangerous situations within a cooperative group of vehicles 182 | 183 | IMM object tracking for high dynamic driving maneuvers 184 | 185 | Switched kalman filterinteracting multiple model algorithm based on optimal autoregressive model for manoeuvring target trackin 186 | 187 | Object tracking in urban intersections based on active use of a priori knowledge: Active interacting multi model filter 188 | 189 | A method for connected vehicle trajectory prediction and collision warning algorithm based on v2v communication 190 | 191 | Interaction-aware motion prediction for autonomous driving:Amultiple model Kalman filtering scheme 192 | 193 | ##### Monte Carlo 194 | 195 | Monte Carlo road safety reasoning 196 | 197 | Comparison of Markov chain abstraction and monte carlo simulation for the safety assessment of autonomous cars 198 | 199 | Driver intention-based vehicle threat assessment using random forests and particle filtering 200 | 201 | Trajectory planning and safety assessment of autonomous vehicles based onmotion prediction and model predictive control 202 | 203 | #### Machine learning 204 | 205 | ##### **Gaussian Process** 206 | 207 | A Bayesian nonparametric approach to modelingmobility patterns 208 | 209 | A Bayesian nonparametric approach to modeling motion patterns 210 | 211 | Online maneuver recognition and multimodal trajectory prediction for intersection assistance using non-parametric regression 212 | 213 | Probabilistic analysis of dynamic scenes and collision risks assessment to improve driving safety 214 | 215 | Unfreezing the robot: Navigation in dense, interacting crowds 216 | 217 | Modeling multi-vehicle interaction scenarios using Gaussian random field 218 | 219 | Motion prediction for moving objects: A statistical approach 220 | 221 | Long-term vehicle motion prediction 222 | 223 | ##### **SVM** 224 | 225 | Using support vector machines for lane-change detection 226 | 227 | Learning-based approach for online lane change intention prediction 228 | 229 | Using support vectormachines and Bayesian filtering for classifying agent intentions at road intersections 230 | 231 | Threat assessment design for driver assistance system at intersections 232 | 233 | ##### **Hidden Markov Model** 234 | 235 | Probabilistic analysis of dynamic scenes and collision risks assessment to improve driving safety 236 | 237 | Next place prediction using mobility Markov chains 238 | 239 | Improved driving behaviors prediction based on fuzzy logic-hidden Markov model (fl-hmm) 240 | 241 | Continuous driver intention recognition with hiddenMarkov models 242 | 243 | A self-adaptive parameter selection trajectory prediction approach via hidden Markov models 244 | 245 | Decision-making and planning method for autonomous vehicles based on motivation and risk assessment 246 | 247 | How would surround vehicles move? A unified framework for maneuver classification and motion prediction 248 | 249 | Research on traffic vehicle behavior prediction method based on game theory and hmm 250 | 251 | ##### **Dynamic Bayesian Network** 252 | 253 | Learning driver behavior models from traffic observations for decision making and planning 254 | 255 | An integrated approach to maneuver-based trajectory prediction and criticality assessment in arbitrary road environments 256 | 257 | A game-theoretic approach to replanning-aware interactive scene prediction and planning 258 | 259 | Probabilistic intention prediction and trajectory generation based on dynamic Bayesian networks 260 | 261 | A dynamic Bayesian network for vehicle maneuver prediction in highway driving scenarios: Framework and verification 262 | 263 | Pedestrian trajectory prediction combining probabilistic reasoning and sequence learning 264 | 265 | #### Deep Learning 266 | 267 | ##### Sequence Network 268 | 269 | A recurrent neural network solution for predicting driver intention at unsignalized intersections 270 | 271 | Long short termmemory for driver intent prediction 272 | 273 | Generalizable intention prediction of human drivers at intersections 274 | 275 | An LSTM network for highway trajectory prediction 276 | 277 | Online vehicle trajectory prediction using policy anticipation network and optimization-based context reasoning 278 | 279 | Naturalistic driver intention and path prediction using recurrent neural networks 280 | 281 | Sequence-to-sequence prediction of vehicle trajectory via lstm encoder-decoder architecture 282 | 283 | Personalized vehicle trajectory prediction based on joint time-series modeling for connected vehicles 284 | 285 | Argoverse: 3D tracking and forecasting with rich maps 286 | 287 | Multimodal trajectory predictions for urban environments using geometric relationships between a vehicle and lanes 288 | 289 | Modeling vehicle interactions via modified lstm models for trajectory prediction 290 | 291 | Predicting vehicle behaviors over an extended horizon using behavior interaction network 292 | 293 | Intention-aware long horizon trajectory prediction of surrounding vehicles using dual lstm networks 294 | 295 | Multi-modal trajectory prediction of surrounding vehicles with maneuver based LSTMS 296 | 297 | Rnn-based path prediction of obstacle vehicles with deep ensemble 298 | 299 | Multiple futures prediction 300 | 301 | A recurrent attention and interaction model for pedestrian trajectory prediction 302 | 303 | Vehicle motion prediction at intersections based on the turning intention and prior trajectories model 304 | 305 | Convolutional neural network for trajectory prediction 306 | 307 | Covernet: Multimodal behavior prediction using trajectory sets 308 | 309 | Deep kinematic models for kinematically feasible vehicle trajectory predictions 310 | 311 | Multimodal trajectory predictions for autonomous driving using deep convolutional networks 312 | 313 | Predicting motion of vulnerable road users using highdefinition maps and efficient convnets 314 | 315 | Uncertainty-aware short-term motion prediction of traffic actors for autonomous driving 316 | 317 | Multiple trajectory prediction with deep temporal and spatial convolutional neural networks 318 | 319 | A lane-changing prediction method based on temporal convolution network 320 | 321 | Mantra: Memory augmented networks for multiple trajectory prediction 322 | 323 | Home: Heatmap output for future motion estimation 324 | 325 | Tpcn: Temporal point cloud networks for motion forecasting 326 | 327 | Convolutional social pooling for vehicle trajectory prediction 328 | 329 | Traphic: Trajectory prediction in dense and heterogeneous traffic using weighted interactions 330 | 331 | Motion trajectory prediction based on a CNN-LSTM sequential model 332 | 333 | The importance of prior knowledge in precise multimodal prediction 334 | 335 | Desire: Distant future prediction in dynamic scenes with interacting agents 336 | 337 | Rules of the road: Predicting driving behavior with a convolutional model of semantic interactions 338 | 339 | Multipath: Multiple probabilistic anchor trajectory hypotheses for behavior prediction 340 | 341 | Multi-head attention based probabilistic vehicle trajectory prediction 342 | Attention based vehicle trajectory prediction 343 | 344 | Trajectory prediction for autonomous driving based on multi-head attention with joint agent-map representatio 345 | 346 | Transformer networks for trajectory forecasting 347 | 348 | Multi-modal motion prediction with transformer-based neural network for autonomous driving 349 | 350 | End-to-end contextual perception and prediction with interaction transformer 351 | 352 | Scene transformer:A unified multi-task model for behavior prediction and planning 353 | 354 | Multimodal motion prediction with stacked transformers 355 | 356 | Trajectron: Dynamically-feasible trajectory forecasting with heterogeneous data 357 | 358 | ##### **GNN** 359 | 360 | Graph neural networks for modelling traffic participant interaction 361 | 362 | Grip: Graph-based interaction-aware trajectory prediction 363 | 364 | Grip++: Enhanced graph-based interaction-aware trajectory prediction for autonomous driving 365 | SCALE-Net: Scalable vehicle trajectory prediction network under random number of interacting vehicles via edge-enhanced graph convolutional neural network 366 | 367 | Social-STGCNN: A social spatio-temporal graph convolutional neural network for human trajectory prediction 368 | 369 | Forecasting trajectory and behavior of road-agents using spectral clustering in graph-LSTMS 370 | 371 | Gisnet: Graph-based information sharing network for vehicle trajectory prediction 372 | 373 | Vectornet: Encoding hd maps and agent dynamics from vectorized representation 374 | 375 | Learning lane graph representations for motion forecasting 376 | 377 | Tnt: Target-driven trajectory prediction 378 | 379 | Densetnt: End-to-end trajectory prediction from dense goal sets 380 | 381 | Lanercnn: Distributed representations for graph-centric motion forecasting 382 | 383 | Stgat: Modeling spatialtemporal interactions for human trajectory prediction 384 | 385 | Stochastic trajectory predictionwith social graph network 386 | 387 | ##### Generative Model 388 | 389 | Learning to predict vehicle trajectories with model-based planning 390 | 391 | Sequence-to-sequence prediction of vehicle trajectory via lstm encoder decoder architecture 392 | 393 | Multi-modal trajectory prediction of surrounding vehicles with maneuver based LSTMS 394 | 395 | Multiple futures prediction 396 | 397 | Rules of the road: Predicting driving behavior with a convolutional model of semantic interactions 398 | 399 | Social gan: Socially acceptable trajectories with generative adversarial networks 400 | 401 | Tppo: A novel trajectory predictor with pseudo oracle 402 | 403 | Conditional generative neural system for probabilistic trajectory prediction 404 | 405 | Sophie: An attentive gan for predicting paths compliant to social and physical constraints 406 | 407 | Vehicle trajectory prediction using gan 408 | 409 | Multi-agent tensor fusion for contextual trajectory prediction 410 | 411 | Multi-vehicle collaborative learning for trajectory prediction with spatio-temporal tensor fusion 412 | 413 | R2p2: A reparameterized pushforward policy for diverse, precise generative path forecasting 414 | 415 | Implicit latent variable model for scene-consistent motion forecasting 416 | 417 | 418 | 419 | #### Reinforcement Learning 420 | 421 | ##### **Inverse Reinforcement Learning** 422 | 423 | Maximum margin planning 424 | 425 | Inverse reinforcement learning through structured classification 426 | 427 | Learning autonomous driving styles and maneuvers from expert demonstration 428 | 429 | Maximum entropy inverse reinforcement learning 430 | 431 | Learning to drive using inverse reinforcement learning and deep q-networks 432 | 433 | Probabilistic prediction of interactive driving behavior via hierarchical inverse reinforcement learning 434 | 435 | Modeling driver behavior from demonstrations in dynamic environments using spatiotemporal lattices 436 | 437 | Learning from naturalistic driving data for human-like autonomous highway driving 438 | 439 | Efficient samplingbased maximum entropy inverse reinforcement learning with application to autonomous driving 440 | 441 | Trajectory forecasts in unknown environments conditioned on grid-based plans 442 | 443 | Accelerated inverse reinforcement learning with randomly pre-sampled policies for autonomous driving reward design 444 | 445 | Learning trajectory prediction with continuous inverse optimal control via langevin sampling of energy-based models 446 | 447 | Driving behavior modeling using naturalistic human driving data with inverse reinforcement learning 448 | 449 | ##### **Generative Adversarial Imitation Learning** 450 | 451 | Imitating driver behavior with generative adversarial networks 452 | 453 | Infogail: Interpretable imitation learning from visual demonstrations 454 | 455 | Modeling human driving behavior through generative adversarial imitation learning 456 | 457 | Trajgail: Generating urban vehicle trajectories using generative adversarial imitation learning 458 | 459 | ##### **Deep Inverse Reinforcement Learning** 460 | 461 | Deep inverse reinforcement learning for behavior prediction in autonomous driving: Accurate forecasts of vehicle motion 462 | 463 | Advanced planning for autonomous vehicles using reinforcement learning and deep inverse reinforcement learning 464 | 465 | Large-scale cost function learning for path planning using deep inverse reinforcement learning 466 | 467 | Inverse reinforcement learning via neural network in driver behavior modeling 468 | 469 | Off-road autonomous vehicles traversability analysis and trajectory planning based on deep inverse reinforcement learning 470 | 471 | Incorporating multi-context into the traversability map for urban autonomous driving using deep inverse reinforcement learning 472 | 473 | ### 2022 Latest Trajectory Forecast Survey 474 | 475 | A Survey on Trajectory-Prediction Methods for Autonomous Driving 476 | 477 | 478 | 479 | 480 | 481 | ## 四、Methods 482 | 483 | ### CVPR 2023 484 | 485 | - Decompose More and Aggregate Better: Two Closer Looks at Frequency Representation Learning for Human Motion Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Gao_Decompose_More_and_Aggregate_Better_Two_Closer_Looks_at_Frequency_CVPR_2023_paper.pdf)** 486 | - DeFeeNet: Consecutive 3D Human Motion Prediction with Deviation Feedback. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Sun_DeFeeNet_Consecutive_3D_Human_Motion_Prediction_With_Deviation_Feedback_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2304.04496)** 487 | - EqMotion: Equivariant Multi-agent Motion Prediction with Invariant Interaction Reasoning. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Xu_EqMotion_Equivariant_Multi-Agent_Motion_Prediction_With_Invariant_Interaction_Reasoning_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2303.10876)** **[Code](https://github.com/MediaBrain-SJTU/EqMotion)** 488 | - FEND: A Future Enhanced Distribution-Aware Contrastive Learning Framework for Long-tail Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Wang_FEND_A_Future_Enhanced_Distribution-Aware_Contrastive_Learning_Framework_for_Long-Tail_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2303.16574)** **[Code](https://github.com/RLuke22/FJMP)** **[Website](https://rluke22.github.io/FJMP/)** 489 | - FJMP: Factorized Joint Multi-Agent Motion Prediction over Learned Directed Acyclic Interaction Graphs. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Rowe_FJMP_Factorized_Joint_Multi-Agent_Motion_Prediction_Over_Learned_Directed_Acyclic_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2211.16197)** 490 | - IPCC-TP: Utilizing Incremental Pearson Correlation Coefficient for Joint Multi-Agent Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Zhu_IPCC-TP_Utilizing_Incremental_Pearson_Correlation_Coefficient_for_Joint_Multi-Agent_Trajectory_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2303.00575)** 491 | - Leapfrog Diffusion Model for Stochastic Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Mao_Leapfrog_Diffusion_Model_for_Stochastic_Trajectory_Prediction_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2303.10895)** **[Code](https://github.com/MediaBrain-SJTU/LED)** 492 | - MotionDiffuser: Controllable Multi-Agent Motion Prediction using Diffusion. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Jiang_MotionDiffuser_Controllable_Multi-Agent_Motion_Prediction_Using_Diffusion_CVPR_2023_paper.pdf)** 493 | - Planning-oriented Autonomous Driving. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Hu_Planning-Oriented_Autonomous_Driving_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2212.10156)** **[Code](https://github.com/OpenDriveLab/UniAD)** **[Website](https://opendrivelab.github.io/UniAD/)** 494 | - ProphNet: Efficient Agent-Centric Motion Forecasting with Anchor-Informed Proposals. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Wang_ProphNet_Efficient_Agent-Centric_Motion_Forecasting_With_Anchor-Informed_Proposals_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2303.12071)** 495 | - Query-Centric Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Zhou_Query-Centric_Trajectory_Prediction_CVPR_2023_paper.pdf)** 496 | - Stimulus Verification is a Universal and Effective Sampler in Multi-modal Human Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Sun_Stimulus_Verification_Is_a_Universal_and_Effective_Sampler_in_Multi-Modal_CVPR_2023_paper.pdf)** 497 | - Trace and Pace: Controllable Pedestrian Animation via Guided Trajectory Diffusion. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Rempe_Trace_and_Pace_Controllable_Pedestrian_Animation_via_Guided_Trajectory_Diffusion_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2304.01893)** **[Website](https://nv-tlabs.github.io/trace-pace/)** 498 | - Trajectory-Aware Body Interaction Transformer for Multi-Person Pose Forecasting. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Peng_Trajectory-Aware_Body_Interaction_Transformer_for_Multi-Person_Pose_Forecasting_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2303.05095)** **[Code](https://github.com/xiaogangpeng/TBIFormer)** 499 | - Uncovering the Missing Pattern: Unified Framework Towards Trajectory Imputation and Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Xu_Uncovering_the_Missing_Pattern_Unified_Framework_Towards_Trajectory_Imputation_and_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2303.16005)** **[Code](https://github.com/colorfulfuture/GC-VRNN)** 500 | - Unsupervised Sampling Promoting for Stochastic Human Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Chen_Unsupervised_Sampling_Promoting_for_Stochastic_Human_Trajectory_Prediction_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2304.04298)** **[Code](https://github.com/viewsetting/Unsupervised_sampling_promoting)** 501 | - ViP3D: End-to-end Visual Trajectory Prediction via 3D Agent Queries. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Gu_ViP3D_End-to-End_Visual_Trajectory_Prediction_via_3D_Agent_Queries_CVPR_2023_paper.pdf)** **[arXiv](https://arxiv.org/abs/2208.01582)** **[Website](https://tsinghua-mars-lab.github.io/ViP3D/)** 502 | - Weakly Supervised Class-agnostic Motion Prediction for Autonomous Driving. **[Paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Li_Weakly_Supervised_Class-Agnostic_Motion_Prediction_for_Autonomous_Driving_CVPR_2023_paper.pdf)** 503 | 504 | ### ICLR 2023 505 | 506 | - Learning Heterogeneous Interaction Strengths by Trajectory Prediction with Graph Neural Network. **[OpenReview](https://openreview.net/forum?id=qU6NIcpaSi-)** **[arXiv](https://arxiv.org/abs/2208.13179)** 507 | - Stochastic Multi-Person 3D Motion Forecasting. **[OpenReview](https://openreview.net/forum?id=_s1N-DnxdyT)** 508 | 509 | ### ICRA 2023 510 | 511 | - Expanding the Deployment Envelope of Behavior Prediction via Adaptive Meta-Learning. **[arXiv](https://arxiv.org/abs/2209.11820)** 512 | 513 | ### arXiv 2022 514 | 515 | - Safety-compliant Generative Adversarial Networks for Human Trajectory Forecasting. **[arXiv](https://arxiv.org/abs/2209.12243)** 516 | - Wayformer: Motion Forecasting via Simple & Efficient Attention Networks. **[arXiv](https://arxiv.org/abs/2207.05844)** 517 | 518 | ### CoRL 2022 519 | 520 | - SSL-Lanes: Self-Supervised Learning for Motion Forecasting in Autonomous Driving. **[arXiv](https://arxiv.org/abs/2206.14116)** **[Code](https://github.com/AutoVision-cloud/SSL-Lanes)** 521 | 522 | ### NeurIPS 2022 523 | 524 | - Contact-aware Human Motion Forecasting. **[OpenReview](https://openreview.net/forum?id=LIKlL1Br9AT)** **[arXiv](https://arxiv.org/abs/2210.03954)** **[Code](https://github.com/wei-mao-2019/ContAwareMotionPred)** 525 | - Forecasting Human Trajectory from Scene History. **[OpenReview](https://openreview.net/forum?id=RW-OOBU11xl)** **[arXiv](https://arxiv.org/abs/2210.08732)** 526 | - Interaction Modeling with Multiplex Attention. **[OpenReview](https://openreview.net/forum?id=SeHslYhFx5-)** **[arXiv](https://arxiv.org/abs/2208.10660)** 527 | - Motion Forecasting Transformer with Global Intention Localization and Local Movement Refinement. **[OpenReview](https://openreview.net/forum?id=9t-j3xDm7_Q)** **[arXiv](https://arxiv.org/abs/2209.13508)** **[Code](https://github.com/sshaoshuai/MTR)** 528 | 529 | ### ECCV 2022 530 | 531 | - Action-based Contrastive Learning for Trajectory Prediction. **[arXiv](https://arxiv.org/abs/2207.08664)** 532 | - AdvDO: Realistic Adversarial Attacks for Trajectory Prediction. **[arXiv](https://arxiv.org/abs/2209.08744)** 533 | - Aware of the History: Trajectory Forecasting with the Local Behavior Data. **[arXiv](https://arxiv.org/abs/2207.09646)** **[Code](https://github.com/Kay1794/LocalBehavior-based-trajectory-prediction)** 534 | - Diverse Human Motion Prediction Guided by Multi-Level Spatial-Temporal Anchors **[arXiv](https://arxiv.org/abs/2302.04860)** **[Code](https://github.com/Sirui-Xu/STARS)** **[Website](https://sirui-xu.github.io/STARS/)** 535 | - D2-TPred: Discontinuous Dependency for Trajectory Prediction under Traffic Lights. **[arXiv](https://arxiv.org/abs/2207.10398)** **[Code](https://github.com/VTP-TL/D2-TPred)** 536 | - Entry-Flipped Transformer for Inference and Prediction of Participant Behavior. **[arXiv](https://arxiv.org/abs/2207.06235)** 537 | - Hierarchical Latent Structure for Multi-Modal Vehicle Trajectory Forecasting. **[arXiv](https://arxiv.org/abs/2207.04624)** 538 | - Human Trajectory Prediction via Neural Social Physics. **[arXiv](https://arxiv.org/abs/2207.10435)** **[Code](https://github.com/realcrane/Human-Trajectory-Prediction-via-Neural-Social-Physics)** 539 | - Learning Pedestrian Group Representations for Multi-modal Trajectory Prediction. **[arXiv](https://arxiv.org/abs/2207.09953)** **[Code](https://github.com/inhwanbae/GPGraph)** **[Website](https://inhwanbae.github.io/publication/gpgraph/#)** 540 | - Overlooked Poses Actually Make Sense: Distilling Privileged Knowledge for Human Motion Prediction. **[arXiv](https://arxiv.org/abs/2208.01302)** 541 | - Polarimetric Pose Prediction. **[arXiv](https://arxiv.org/abs/2112.03810)** 542 | - PreTraM: Self-Supervised Pre-training via Connecting Trajectory and Map. **[arXiv](https://arxiv.org/abs/2204.10435)** **[Code](https://github.com/chenfengxu714/PreTraM)** 543 | - Skeleton-Parted Graph Scattering Networks for 3D Human Motion Prediction. **[arXiv](https://arxiv.org/abs/2208.00368)** **[Code](https://github.com/MediaBrain-SJTU/SPGSN)** 544 | - Social-Implicit: Rethinking Trajectory Prediction Evaluation and The Effectiveness of Implicit Maximum Likelihood Estimatio. **[arXiv](https://arxiv.org/abs/2203.03057)** **[Code](https://github.com/abduallahmohamed/Social-Implicit)** **[Website](https://www.abduallahmohamed.com/social-implicit-amdamv-adefde-demo)** 545 | - Social ODE: Multi-Agent Trajectory Forecasting with Neural Ordinary Differential Equations. 546 | - Social-SSL: Self-Supervised Cross-Sequence Representation Learning Based on Transformers for Multi-Agent Trajectory Prediction. **[Paper](https://basiclab.lab.nycu.edu.tw/assets/Social-SSL.pdf)** **[Code](https://github.com/Sigta678/Social-SSL)** 547 | - SocialVAE: Human Trajectory Prediction using Timewise Latents. **[arXiv](https://arxiv.org/abs/2203.08207)** **[Code](https://github.com/xupei0610/SocialVAE)** 548 | - ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning. **[arXiv](https://arxiv.org/abs/2207.07601)** **[Code](https://github.com/OpenPerceptionX/ST-P3)** 549 | - View Vertically: A Hierarchical Network for Trajectory Prediction via Fourier Spectrums. **[arXiv](https://arxiv.org/abs/2110.07288)** 550 | 551 | ### CVPR 2022 552 | 553 | #### Trajectory Prediction Related 554 | 555 | - Adaptive Trajectory Prediction via Transferable GNN. **[arXiv](https://arxiv.org/abs/2203.05046)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_Adaptive_Trajectory_Prediction_via_Transferable_GNN_CVPR_2022_paper.pdf)** 556 | - ATPFL: Automatic Trajectory Prediction Model Design Under Federated Learning Framework. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_ATPFL_Automatic_Trajectory_Prediction_Model_Design_Under_Federated_Learning_Framework_CVPR_2022_paper.pdf)** 557 | - Convolutions for Spatial Interaction Modeling. **[arXiv](https://arxiv.org/abs/2104.07182)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Su_Convolutions_for_Spatial_Interaction_Modeling_CVPR_2022_paper.pdf)** 558 | - End-to-End Trajectory Distribution Prediction Based on Occupancy Grid Maps. **[arXiv](https://arxiv.org/abs/2203.16910)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Guo_End-to-End_Trajectory_Distribution_Prediction_Based_on_Occupancy_Grid_Maps_CVPR_2022_paper.pdf)** **[Code](https://github.com/Kguo-cs/TDOR)** 559 | - Forecasting from LiDAR via Future Object Detection. **[arXiv](https://arxiv.org/abs/2203.16297)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Peri_Forecasting_From_LiDAR_via_Future_Object_Detection_CVPR_2022_paper.pdf)** **[Code](https://github.com/neeharperi/FutureDet)** 560 | - Graph-based Spatial Transformer with Memory Replay for Multi-future Pedestrian Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Graph-Based_Spatial_Transformer_With_Memory_Replay_for_Multi-Future_Pedestrian_Trajectory_CVPR_2022_paper.pdf)** **[Code](https://github.com/Jacobieee/ST-MR)** 561 | - GroupNet: Multiscale Hypergraph Neural Networks for Trajectory Prediction with Relational Reasoning. **[arXiv](https://arxiv.org/abs/2204.08770)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_GroupNet_Multiscale_Hypergraph_Neural_Networks_for_Trajectory_Prediction_With_Relational_CVPR_2022_paper.pdf)** **[Code](https://github.com/MediaBrain-SJTU/GroupNet)** 562 | - How Many Observations are Enough? Knowledge Distillation for Trajectory Forecasting. **[arXiv](https://arxiv.org/abs/2203.04781)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Monti_How_Many_Observations_Are_Enough_Knowledge_Distillation_for_Trajectory_Forecasting_CVPR_2022_paper.pdf)** 563 | - LTP: Lane-Based Trajectory Prediction for Autonomous Driving. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_LTP_Lane-Based_Trajectory_Prediction_for_Autonomous_Driving_CVPR_2022_paper.pdf)** 564 | - M2I: From Factored Marginal Trajectory Prediction to Interactive Prediction. **[arXiv](https://arxiv.org/abs/2202.11884)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Sun_M2I_From_Factored_Marginal_Trajectory_Prediction_to_Interactive_Prediction_CVPR_2022_paper.pdf)** **[Code](https://github.com/Tsinghua-MARS-Lab/M2I)** 565 | - Neural Prior for Trajectory Estimation. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Neural_Prior_for_Trajectory_Estimation_CVPR_2022_paper.pdf)** **[Website](https://mightychaos.github.io/projects/cvpr22/supplementary/supp.html)** 566 | - Non-Probability Sampling Network for Stochastic Human Trajectory Prediction. **[arXiv](https://arxiv.org/abs/2203.13471)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Bae_Non-Probability_Sampling_Network_for_Stochastic_Human_Trajectory_Prediction_CVPR_2022_paper.pdf)** **[Code](https://github.com/inhwanbae/NPSN)** 567 | - On Adversarial Robustness of Trajectory Prediction for Autonomous Vehicles. **[arXiv](https://arxiv.org/abs/2201.05057)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhang_On_Adversarial_Robustness_of_Trajectory_Prediction_for_Autonomous_Vehicles_CVPR_2022_paper.pdf)** **[Code](https://github.com/zqzqz/AdvTrajectoryPrediction)** 568 | - Remember Intentions: Retrospective-Memory-based Trajectory Prediction. **[arXiv](https://arxiv.org/abs/2203.11474)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_Remember_Intentions_Retrospective-Memory-Based_Trajectory_Prediction_CVPR_2022_paper.pdf)** **[Code](https://github.com/MediaBrain-SJTU/MemoNet)** 569 | - ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_ScePT_Scene-Consistent_Policy-Based_Trajectory_Predictions_for_Planning_CVPR_2022_paper.pdf)** **[Code](https://github.com/nvr-avg/ScePT)** 570 | - Stochastic Trajectory Prediction via Motion Indeterminacy Diffusion. **[arXiv](https://arxiv.org/abs/2203.13777)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Gu_Stochastic_Trajectory_Prediction_via_Motion_Indeterminacy_Diffusion_CVPR_2022_paper.pdf)** **[Code](https://github.com/gutianpei/MID)** 571 | - Vehicle Trajectory Prediction Works, But Not Everywhere. **[arXiv](https://arxiv.org/abs/2112.03909)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Bahari_Vehicle_Trajectory_Prediction_Works_but_Not_Everywhere_CVPR_2022_paper.pdf)** **[Code](https://github.com/vita-epfl/s-attack)** 572 | - Whose Track Is It Anyway? Improving Robustness to Tracking Errors With Affinity-Based Trajectory Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Weng_Whose_Track_Is_It_Anyway_Improving_Robustness_to_Tracking_Errors_CVPR_2022_paper.pdf)** **[Code](https://xinshuoweng.com/projects/Affinipred/)** 573 | - Goal-driven Self-Attentive Recurrent Networks for Trajectory Prediction. (CVPR'22 Workshop Precognition: Seeing Through the Future) **[arXiv](https://arxiv.org/abs/2204.11561)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022W/Precognition/papers/Chiara_Goal-Driven_Self-Attentive_Recurrent_Networks_for_Trajectory_Prediction_CVPRW_2022_paper.pdf)** 574 | - Importance Is in Your Attention: Agent Importance Prediction for Autonomous Driving. (CVPR'22 Workshop Precognition: Seeing Through the Future) **[arXiv](https://arxiv.org/abs/2204.09121)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022W/Precognition/papers/Hazard_Importance_Is_in_Your_Attention_Agent_Importance_Prediction_for_Autonomous_CVPRW_2022_paper.pdf)** 575 | 576 | #### Motion Prediction Related 577 | 578 | - BE-STI: Spatial-Temporal Integrated Network for Class-Agnostic Motion Prediction With Bidirectional Enhancement. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_BE-STI_Spatial-Temporal_Integrated_Network_for_Class-Agnostic_Motion_Prediction_With_Bidirectional_CVPR_2022_paper.pdf)** **[Code](https://github.com/be-sti/be-sti)** 579 | - Forecasting Characteristic 3D Poses of Human Actions. **[arXiv](https://arxiv.org/abs/2011.15079)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Diller_Forecasting_Characteristic_3D_Poses_of_Human_Actions_CVPR_2022_paper.pdf)** **[Code](https://github.com/chrdiller/characteristic3dposes)** 580 | - HiVT: Hierarchical Vector Transformer for Multi-Agent Motion Prediction. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhou_HiVT_Hierarchical_Vector_Transformer_for_Multi-Agent_Motion_Prediction_CVPR_2022_paper.pdf)** **[Code](https://github.com/ZikangZhou/HiVT)** 581 | - Human Trajectory Prediction With Momentary Observation. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Sun_Human_Trajectory_Prediction_With_Momentary_Observation_CVPR_2022_paper.pdf)** 582 | - MotionAug: Augmentation With Physical Correction for Human Motion Prediction. **[arXiv](https://arxiv.org/abs/2203.09116)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Maeda_MotionAug_Augmentation_With_Physical_Correction_for_Human_Motion_Prediction_CVPR_2022_paper.pdf)** **[Code](https://github.com/meaten/MotionAug)** 583 | - Motron: Multimodal Probabilistic Human Motion Forecasting. **[arXiv](https://arxiv.org/abs/2203.04132)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Salzmann_Motron_Multimodal_Probabilistic_Human_Motion_Forecasting_CVPR_2022_paper.pdf)** **[Code](https://github.com/TUM-AAS/motron-cvpr22)** 584 | - Multi-Objective Diverse Human Motion Prediction With Knowledge Distillation. **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Ma_Multi-Objective_Diverse_Human_Motion_Prediction_With_Knowledge_Distillation_CVPR_2022_paper.pdf)** 585 | - Multi-Person Extreme Motion Prediction. **[arXiv](https://arxiv.org/abs/2105.08825)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Guo_Multi-Person_Extreme_Motion_Prediction_CVPR_2022_paper.pdf)** **[Code](https://github.com/GUO-W/MultiMotion)** 586 | - Progressively Generating Better Initial Guesses Towards Next Stages for High-Quality Human Motion Prediction. **[arXiv](https://arxiv.org/abs/2203.16051)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Ma_Progressively_Generating_Better_Initial_Guesses_Towards_Next_Stages_for_High-Quality_CVPR_2022_paper.pdf)** **[Code](https://github.com/705062791/PGBIG)** 587 | - Spatial-Temporal Gating-Adjacency GCN for Human Motion Prediction. **[arXiv](https://arxiv.org/abs/2203.01474)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhong_Spatio-Temporal_Gating-Adjacency_GCN_for_Human_Motion_Prediction_CVPR_2022_paper.pdf)** 588 | - Towards Robust and Adaptive Motion Forecasting: A Causal Representation Perspective. **[arXiv](https://arxiv.org/abs/2111.14820)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Towards_Robust_and_Adaptive_Motion_Forecasting_A_Causal_Representation_Perspective_CVPR_2022_paper.pdf)** **[Code](https://github.com/vita-epfl/causalmotion)** 589 | - Weakly-Supervised Action Transition Learning for Stochastic Human Motion Prediction. **[arXiv](https://arxiv.org/abs/2205.15608)** **[Paper](https://openaccess.thecvf.com/content/CVPR2022/papers/Mao_Weakly-Supervised_Action_Transition_Learning_for_Stochastic_Human_Motion_Prediction_CVPR_2022_paper.pdf)** **[Code](https://github.com/wei-mao-2019/WAT)** 590 | 591 | ### ICRA 2022 592 | 593 | - Path-Aware Graph Attention for HD Maps in Motion Prediction. **[arXiv](https://arxiv.org/abs/2202.13772)** 594 | 595 | ### ICLR 2022 596 | 597 | - D-CODE: Discovering Closed-form ODEs from Observed Trajectories. **[Paper](https://openreview.net/forum?id=wENMvIsxNN)** **[Code](https://github.com/ZhaozhiQIAN/D-CODE-ICLR-2022)** 598 | - Latent Variable Sequential Set Transformers For Joint Multi-Agent Motion Prediction. **[Paper](https://openreview.net/forum?id=Dup_dDqkZC5)** **[Code](https://gist.github.com/fgolemo/e6ff3daddcf735e8835789bbb39ece58)** 599 | - ProtoRes: Proto-Residual Network for Pose Authoring via Learned Inverse Kinematics. **[Paper](https://openreview.net/forum?id=s03AQxehtd_)** **[Website](https://unity-technologies.github.io/Labs/protores.html)** 600 | - Scene Transformer: A Unified Architecture for Predicting Multiple Agent Trajectories. **[Paper](https://openreview.net/forum?id=Wm3EA5OlHsG)** 601 | - THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling. **[Paper](https://openreview.net/forum?id=QDdJhACYrlX)** 602 | - You Mostly Walk Alone: Analyzing Feature Attribution in Trajectory Prediction. **[Paper](https://openreview.net/forum?id=POxF-LEqnF)** 603 | --------------------------------------------------------------------------------