We will update the following list after the paper is accepted.
- [2025-02-07] We have released our project page.
- We have uploaded our paper, NeuralGS on arXiv!
- Upload the code.
Table 1. Quantitative results evaluated on Mip-NeRF 360, Tanks&Temples, and Deep Blending datasets. We highlight the best-performing results in red and the second-best results in yellow for all compression methods
Table 2. Quantitative results of the proposed method evaluated on the NeRF-Synthetic dataset. We highlight the best-performing results in red and the second-best results in yellow for all compression methods. |
Table 3. Performance comparison with 3DGS. Rendering FPS and model size (MB) are reported. The rendering speed of both methods is measured on our machine. 
Table 4. Quantitative ablation study on the Deep Blending dataset by progressively adding our proposed improvement. |
This source code is derived from multiple sources, in particular: gaussian-splatting. We thank the authors for releasing their code.
@misc{tang2025neuralgsbridgingneuralfields,
title={NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations},
author={Zhenyu Tang and Chaoran Feng and Xinhua Cheng and Wangbo Yu and Junwu Zhang and Yuan Liu and Xiaoxiao Long and Wenping Wang and Li Yuan},
year={2025},
eprint={2503.23162},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2503.23162},
}