Efficient Training of Neural Fields

This repository contains efficient training schemes for neural fields, developed by EffL Group at POSTECH.

Paper List

[CVPR 2024, Oral] In Search of a Data Transformation That Accelerates Neural Field Training
by Junwon Seo*, Sangyoon Lee*, Kwang In Kim, and Jaeho Lee [Paper], [Hugging Face Demo]

[ICLR 2025] Fast Training of Sinusoidal Neural Fields via Scaling Initialization
by Taesun Yeom*, Sangyoon Lee*, and Jaeho Lee [Paper]

Experiments

Please refer to the README file in each folder for instructions on running the code and for more details.

Citation

@inproceedings{RPP,
author = {Junwon Seo and Sangyoon Lee and Kwang In Kim and Jaeho Lee},
title = {In Search of a Data Transformation That Accelerates Neural Field Training},
year = {2024},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}
}

@inproceedings{
yeom2025fast,
title={Fast Training of Sinusoidal Neural Fields via Scaling Initialization},
author={Taesun Yeom and Sangyoon Lee and Jaeho Lee},
booktitle={The Thirteenth International Conference on Learning Representations},
year={2025},
url={https://openreview.net/forum?id=Sr5XaZzirA}
}