Trajectory Forecasting
73 papers with code • 4 benchmarks • 16 datasets
Trajectory forecasting is a sequential prediction task, where a forecasting model predicts future trajectories of all moving agents (humans, vehicles, etc.) in a scene, based on their past trajectories and/or the scene context.
(Illustrative figure from Social NCE: Contrastive Learning of Socially-aware Motion Representations)
Benchmarks
These leaderboards are used to track progress in Trajectory Forecasting
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
Datasets
UCY
highD Dataset
inD Dataset
JAAD
TrajNet
rounD Dataset
FPL
PIE
exiD Dataset
TrajAir: A General Aviation Trajectory Dataset
Latest papers
KI-GAN: Knowledge-Informed Generative Adversarial Networks for Enhanced Multi-Vehicle Trajectory Forecasting at Signalized Intersections
Reliable prediction of vehicle trajectories at signalized intersections is crucial to urban traffic management and autonomous driving systems.
HPNet: Dynamic Trajectory Forecasting with Historical Prediction Attention
The proposed Historical Prediction Attention together with the Agent Attention and Mode Attention is further formulated as the Triple Factorized Attention module, serving as the core design of HPNet. Experiments on the Argoverse and INTERACTION datasets show that HPNet achieves state-of-the-art performance, and generates accurate and stable future trajectories.
Producing and Leveraging Online Map Uncertainty in Trajectory Prediction
High-definition (HD) maps have played an integral role in the development of modern autonomous vehicle (AV) stacks, albeit with high associated labeling and maintenance costs.
JRDB-Traj: A Dataset and Benchmark for Trajectory Forecasting in Crowds
To address this, we introduce a novel dataset for end-to-end trajectory forecasting, facilitating the evaluation of models in scenarios involving less-than-ideal preceding modules such as tracking.
Fast, Expressive SE$(n)$ Equivariant Networks through Weight-Sharing in Position-Orientation Space
ebekkers/ponita
•
•
The theory of homogeneous spaces tells us how to do group convolutions with feature maps over the homogeneous space of positions $\mathbb{R}^3$, position and orientations $\mathbb{R}^3 {\times} S^2$, and the group $SE(3)$ itself.
Streaming Motion Forecasting for Autonomous Driving
ziqipang/streamingforecasting
•
•
Our benchmark inherently captures the disappearance and re-appearance of agents, presenting the emergent challenge of forecasting for occluded agents, which is a safety-critical problem yet overlooked by snapshot-based benchmarks.
Staged Contact-Aware Global Human Motion Forecasting
l-scofano/stag
•
•
So far, only Mao et al. NeurIPS'22 have addressed scene-aware global motion, cascading the prediction of future scene contact points and the global motion estimation.
TrajPAC: Towards Robustness Verification of Pedestrian Trajectory Prediction Models
zl-helios/trajpac
•
•
ICCV 2023
Firstly, the previous definitions of robustness in trajectory prediction are ambiguous.
trajdata: A Unified Interface to Multiple Human Trajectory Datasets
nvlabs/trajdata
•
•
NeurIPS 2023
The field of trajectory forecasting has grown significantly in recent years, partially owing to the release of numerous large-scale, real-world human trajectory datasets for autonomous vehicles (AVs) and pedestrian motion tracking.
EigenTrajectory: Low-Rank Descriptors for Multi-Modal Trajectory Forecasting
inhwanbae/eigentrajectory
•
•
ICCV 2023
In this paper, we present EigenTrajectory ($\mathbb{ET}$), a trajectory prediction approach that uses a novel trajectory descriptor to form a compact space, known here as $\mathbb{ET}$ space, in place of Euclidean space, for representing pedestrian movements.
