AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data
Abstract
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.
Cite
Text
Zaffaroni et al. "AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data." European Conference on Computer Vision Workshops, 2024. doi:10.1007/978-3-031-91767-7_2Markdown
[Zaffaroni et al. "AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data." European Conference on Computer Vision Workshops, 2024.](https://mlanthology.org/eccvw/2024/zaffaroni2024eccvw-aasgan/) doi:10.1007/978-3-031-91767-7_2BibTeX
@inproceedings{zaffaroni2024eccvw-aasgan,
title = {{AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data}},
author = {Zaffaroni, Mirko and Signoretta, Federico and Grangetto, Marco and Fiandrotti, Attilio},
booktitle = {European Conference on Computer Vision Workshops},
year = {2024},
pages = {15-30},
doi = {10.1007/978-3-031-91767-7_2},
url = {https://mlanthology.org/eccvw/2024/zaffaroni2024eccvw-aasgan/}
}