mlsanityData to 14 Oct 2026

Social-Implicit: Rethinking Trajectory Prediction Evaluation and The Effectiveness of Implicit Maximum Likelihood Estimation

arXiv 2203.03057 ↗First posted 6 Mar 2022cs.CV cs.LG cs.ROcited by 40

Not on the map. Why?

Abstract (arXiv)

Best-of-N (BoN) Average Displacement Error (ADE)/ Final Displacement Error (FDE) is the most used metric for evaluating trajectory prediction models. Yet, the BoN does not quantify the whole generated samples, resulting in an incomplete view of the model's prediction quality and performance. We propose a new metric, Average Mahalanobis Distance (AMD) to tackle this issue. AMD is a metric that quantifies how close the whole generated samples are to the ground truth. We also introduce the Average Maximum Eigenvalue (AMV) metric that quantifies the overall spread of the predictions. Our metrics are validated empirically by showing that the ADE/FDE is not sensitive to distribution shifts, giving a biased sense of accuracy, unlike the AMD/AMV metrics. We introduce the usage of Implicit Maximum Likelihood Estimation (IMLE) as a replacement for traditional generative models to train our model, Social-Implicit. IMLE training mechanism aligns with AMD/AMV objective of predicting trajectories that are close to the ground truth with a tight spread. Social-Implicit is a memory efficient deep model with only 5.8K parameters that runs in real time of about 580Hz and achieves competitive results. Interactive demo of the problem can be seen at https://www.abduallahmohamed.com/social-implicit-amdamv-adefde-demo . Code is available at https://github.com/abduallahmohamed/Social-Implicit .

Key facts. Cited by 40 papers in the map (data through 14 Oct 2026). Builds on 24 of its 45 references that are in the map. First posted on arXiv 6 Mar 2022.

Builds on

24 of 45 references.

References · 24

  1. 2018Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networkscited by 690
  2. 2019MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Predictioncited by 420
  3. 2020Learning Lane Graph Representations for Motion Forecastingcited by 382
  4. 2021AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecastingcited by 331
  5. 2020Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Predictioncited by 325
  6. 2017DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agentscited by 310
  7. 2018SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraintscited by 298
  8. 2018Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networkscited by 289
  9. 2019Human Motion Trajectory Prediction: A Surveycited by 232
  10. 2019Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networkscited by 229
  11. 2018Convolutional Social Pooling for Vehicle Trajectory Predictioncited by 221
  12. 2019PRECOG: PREdiction Conditioned On Goals in Visual Multi-Agent Settingscited by 179
12 more
  1. 2019Multiple Futures Predictioncited by 178
  2. 2020From Goals, Waypoints & Paths To Long Term Human Trajectory Forecastingcited by 167
  3. 2020MotionNet: Joint Perception and Motion Prediction for Autonomous Driving Based on Bird's Eye View Mapscited by 100
  4. 2019Social Ways: Learning Multi-Modal Distributions of Pedestrian Trajectories with GANscited by 79
  5. 2019GRIP++: Enhanced Graph-based Interaction-aware Trajectory Prediction for Autonomous Drivingcited by 73
  6. 2020Social NCE: Contrastive Learning of Socially-aware Motion Representationscited by 58
  7. 2018Implicit Maximum Likelihood Estimationcited by 52
  8. 2021On Exposing the Challenging Long Tail in Future Prediction of Traffic Actorscited by 48
  9. 2019DiversityGAN: Diversity-Aware Vehicle Motion Prediction via Latent Semantic Samplingcited by 40
  10. 2020DAG-Net: Double Attentive Graph Neural Network for Trajectory Forecastingcited by 30
  11. 2019Deep Neural Network Ensembles against Deception: Ensemble Diversity, Accuracy and Robustnesscited by 18
  12. 2016Robust Deep-Learning-Based Road-Prediction for Augmented Reality Navigation Systemscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 12; 2024: 11; 2025: 12; 2026: 5.

40 papers.

Sort:

By area:

  1. 2023EigenTrajectory: Low-Rank Descriptors for Multi-Modal Trajectory Forecastingcited by 41
  2. 2023Fast Inference and Update of Probabilistic Density Estimation on Trajectory Predictioncited by 27
  3. 2024LG-Traj: LLM Guided Pedestrian Trajectory Predictioncited by 25
  4. 2025Unified Spatial-Temporal Edge-Enhanced Graph Networks for Pedestrian Trajectory Predictioncited by 13
  5. 2023Guiding Online Reinforcement Learning with Action-Free Offline Pretrainingcited by 11
  6. 2023Evaluating Pedestrian Trajectory Prediction Methods with Respect to Autonomous Drivingcited by 11
  7. 2023Social Motion Prediction with Cognitive Hierarchiescited by 9
  8. 2025BP-SGCN: Behavioral Pseudo-Label Informed Sparse Graph Convolution Network for Pedestrian and Heterogeneous Trajectory Predictioncited by 9
  9. 2023Vision-based Multi-future Trajectory Prediction: A Surveycited by 8
  10. 2023Best Practices for 2-Body Pose Forecastingcited by 8
  11. 2025Pedestrian Trajectory Prediction Based on Social Interactions Learning With Random Weightscited by 8
  12. 2024Risk-Aware Vehicle Trajectory Prediction Under Safety-Critical Scenarioscited by 7
28 more
  1. 2023Staged Contact-Aware Global Human Motion Forecastingcited by 6
  2. 2024MGF: Mixed Gaussian Flow for Diverse Trajectory Predictioncited by 6
  3. 2024Uncovering the human motion pattern: Pattern Memory-based Diffusion Model for Trajectory Predictioncited by 4
  4. 2024Understanding complex crowd dynamics with generative neural simulatorscited by 4
  5. 2023Discovering Intrinsic Spatial-Temporal Logic Rules to Explain Human Actionscited by 3
  6. 2023Sparse Pedestrian Character Learning for Trajectory Predictioncited by 3
  7. 2024Testing autonomous vehicles and AI: perspectives and challenges from cybersecurity, transparency, robustness and fairnesscited by 3
  8. 2024Context-aware Multi-task Learning for Pedestrian Intent and Trajectory Predictioncited by 3
  9. 2024Snapshot: Towards Application-centered Models for Pedestrian Trajectory Prediction in Urban Traffic Environmentscited by 3
  10. 2025Recent Advances in Multi-Agent Human Trajectory Prediction: A Comprehensive Reviewcited by 2
  11. 2025Uncertainty-aware Probabilistic 3D Human Motion Forecasting via Invertible Networkscited by 2
  12. 2025Floating-Body Hydrodynamic Neural Networkscited by 2
  13. 2023Spatial-Temporal Conditional Random Field for Human Trajectory Predictioncited by 1
  14. 2024Hyper-STTN: Hypergraph Augmented Spatial-Temporal Transformer Network for Trajectory Predictioncited by 1
  15. 2024iMotion-LLM: Instruction-Conditioned Trajectory Generationcited by 1
  16. 2024Pedestrian motion prediction evaluation for urban autonomous drivingcited by 1
  17. 2025An Analysis of Kalman Filter based Object Tracking Methods for Fast-Moving Tiny Objectscited by 1
  18. 2025Reverberation: Learning the Latencies Before Forecasting Trajectoriescited by 1
  19. 2026EgoTraj: Real-World Egocentric Human Trajectory Dataset for Multimodal Predictioncited by 1
  20. 2023FollowMe: Vehicle Behaviour Prediction in Autonomous Vehicle Settingscited by 0
  21. 2025A Spatio-temporal Graph Network Allowing Incomplete Trajectory Input for Pedestrian Trajectory Predictioncited by 0
  22. 2025An Efficient LiDAR-Camera Fusion Network for Multi-Class 3D Dynamic Object Detection and Trajectory Predictioncited by 0
  23. 2025Model Predictive Control for Crowd Navigation via Learning-Based Trajectory Predictioncited by 0
  24. 2025GContextFormer: A global context-aware hybrid multi-head attention approach with scaled additive aggregation for multimodal trajectory predictioncited by 0
  25. 2026Sparse Compositional Flow Matching by geometric assembly from motion primitivescited by 0
  26. 2026AIS-Based Vessel Trajectory Prediction Using Memory-Augmented Neural Networkscited by 0
  27. 2026Gaussian-Mixture Latent Flow for Stochastic 3D Human Motion Predictioncited by 0
  28. 2026MamMA: A Mamba-Based Pedestrian Trajectory Prediction Algorithm Considering Occupancy Map and Pedestrian Awareness Statescited by 0