THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling - Mines Paris Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling

Thomas Gilles
Bogdan Stanciulescu
Fabien Moutarde


In this paper, we propose THOMAS, a joint multi-agent trajectory prediction framework allowing for an efficient and consistent prediction of multi-agent multimodal trajectories. We present a unified model architecture for simultaneous agent future heatmap estimation, in which we leverage hierarchical and sparse image generation for fast and memory-efficient inference. We propose a learnable trajectory recombination model that takes as input a set of predicted trajectories for each agent and outputs its consistent reordered recombination. This recombination module is able to realign the initially independent modalities so that they do no collide and are coherent with each other. We report our results on the Interaction multi-agent prediction challenge and rank 1 st on the online test leaderboard.
Fichier principal
Vignette du fichier
THOMAS_ICLR.pdf (2.81 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03683506 , version 1 (31-05-2022)


  • HAL Id : hal-03683506 , version 1


Thomas Gilles, Stefano Sabatini, Dzmitry Tsishkou, Bogdan Stanciulescu, Fabien Moutarde. THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling. International Conference on Learning Representations, Apr 2022, Virtuel, France. ⟨hal-03683506⟩
53 Consultations
175 Téléchargements


Gmail Facebook X LinkedIn More