Salvato in:
Dettagli Bibliografici
Autori principali: Urano, Yota, Taketsugu, Hiromu, Ukita, Norimichi
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2506.18291
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909656613388288
author Urano, Yota
Taketsugu, Hiromu
Ukita, Norimichi
author_facet Urano, Yota
Taketsugu, Hiromu
Ukita, Norimichi
contents This paper presents an architecture for selecting important neighboring people to predict the primary person's trajectory. To achieve effective neighboring people selection, we propose a people selection module called the Importance Estimator which outputs the importance of each neighboring person for predicting the primary person's future trajectory. To prevent gradients from being blocked by non-differentiable operations when sampling surrounding people based on their importance, we employ the Gumbel Softmax for training. Experiments conducted on the JRDB dataset show that our method speeds up the process with competitive prediction accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18291
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Selective Social-Interaction via Individual Importance for Fast Human Trajectory Prediction
Urano, Yota
Taketsugu, Hiromu
Ukita, Norimichi
Computer Vision and Pattern Recognition
Artificial Intelligence
This paper presents an architecture for selecting important neighboring people to predict the primary person's trajectory. To achieve effective neighboring people selection, we propose a people selection module called the Importance Estimator which outputs the importance of each neighboring person for predicting the primary person's future trajectory. To prevent gradients from being blocked by non-differentiable operations when sampling surrounding people based on their importance, we employ the Gumbel Softmax for training. Experiments conducted on the JRDB dataset show that our method speeds up the process with competitive prediction accuracy.
title Selective Social-Interaction via Individual Importance for Fast Human Trajectory Prediction
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2506.18291