README.md

SEED4D: A Synthetic Ego–Exo Dynamic 4D Driving

WACV 2025

Marius Kästingschäfer and Théo Gieruc and Sebastian Bernhard and Dylan Campbell and Eldar Insafutdinov and Eyvaz Najafli and Thomas Brox

Project page | Paper | Data

Models for egocentric 3D and 4D reconstruction, including few-shot interpolation and extrapolation settings, can benefit from having images from exocentric viewpoints as supervision signals. No existing dataset provides the necessary mixture of complex, dynamic, and multi-view data. To facilitate the development of 3D and 4D reconstruction methods in the autonomous driving context, we propose a Synthetic Ego--Exo Dynamic 4D (SEED4D) dataset. We here provide the code with which we generated the data.

Our data generator is build on CARLA and can generator synthetic egocentrical views similar to the camera setup in nuScenes, KITTI360 or Waymo and simultaneously create create exocentrical views. Resulting poses are outputted in a Nerfstudio suitable data format.

Overview

• Carla Sanity check
• Dataset creation
• Dataset structure
• Dataset visualization

Acknowledgements

We thank the creators of NeRFStudio, and the CARLA simulator for generously open-sourcing their code.

License

Copyright (C) 2025 co-pace GmbH (subsidiary of Continental AG). All rights reserved. This repository is licensed under the BSD-3-Clause license. See LICENSE for the full license text.

Citation

If you find this code useful, please reference in your paper:

@InProceedings{Kastingschafer_2025_WACV,
    author    = {K\"astingsch\"afer, Marius and Gieruc, Th\'eo and Bernhard, Sebastian and Campbell, Dylan and Insafutdinov, Eldar and Najafli, Eyvaz and Brox, Thomas},
    title     = {SEED4D: A Synthetic Ego-Exo Dynamic 4D Data Generator Driving Dataset and Benchmark},
    booktitle = {Proceedings of the Winter Conference on Applications of Computer Vision (WACV)},
    month     = {February},
    year      = {2025},
    pages     = {7741-7753}
}