Saved in:
Bibliographic Details
Main Authors: Zaffaroni, Mirko, Signoretta, Federico, Grangetto, Marco, Fiandrotti, Attilio
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.18038
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912396799377408
author Zaffaroni, Mirko
Signoretta, Federico
Grangetto, Marco
Fiandrotti, Attilio
author_facet Zaffaroni, Mirko
Signoretta, Federico
Grangetto, Marco
Fiandrotti, Attilio
contents Accurately predicting pedestrian trajectories is crucial in applications such as autonomous driving or service robotics, to name a few. Deep generative models achieve top performance in this task, assuming enough labelled trajectories are available for training. To this end, large amounts of synthetically generated, labelled trajectories exist (e.g., generated by video games). However, such trajectories are not meant to represent pedestrian motion realistically and are ineffective at training a predictive model. We propose a method and an architecture to augment synthetic trajectories at training time and with an adversarial approach. We show that trajectory augmentation at training time unleashes significant gains when a state-of-the-art generative model is evaluated over real-world trajectories.
format Preprint
id arxiv_https___arxiv_org_abs_2412_18038
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data
Zaffaroni, Mirko
Signoretta, Federico
Grangetto, Marco
Fiandrotti, Attilio
Computer Vision and Pattern Recognition
Artificial Intelligence
Accurately predicting pedestrian trajectories is crucial in applications such as autonomous driving or service robotics, to name a few. Deep generative models achieve top performance in this task, assuming enough labelled trajectories are available for training. To this end, large amounts of synthetically generated, labelled trajectories exist (e.g., generated by video games). However, such trajectories are not meant to represent pedestrian motion realistically and are ineffective at training a predictive model. We propose a method and an architecture to augment synthetic trajectories at training time and with an adversarial approach. We show that trajectory augmentation at training time unleashes significant gains when a state-of-the-art generative model is evaluated over real-world trajectories.
title AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2412.18038