NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908487295959040 |
|---|---|
| author | Tang, Zhenyu Feng, Chaoran Cheng, Xinhua Yu, Wangbo Zhang, Junwu Liu, Yuan Long, Xiaoxiao Wang, Wenping Yuan, Li |
| author_facet | Tang, Zhenyu Feng, Chaoran Cheng, Xinhua Yu, Wangbo Zhang, Junwu Liu, Yuan Long, Xiaoxiao Wang, Wenping Yuan, Li |
| contents | 3D Gaussian Splatting (3DGS) achieves impressive quality and rendering speed, but with millions of 3D Gaussians and significant storage and transmission costs. In this paper, we aim to develop a simple yet effective method called NeuralGS that compresses the original 3DGS into a compact representation. Our observation is that neural fields like NeRF can represent complex 3D scenes with Multi-Layer Perceptron (MLP) neural networks using only a few megabytes. Thus, NeuralGS effectively adopts the neural field representation to encode the attributes of 3D Gaussians with MLPs, only requiring a small storage size even for a large-scale scene. To achieve this, we adopt a clustering strategy and fit the Gaussians within each cluster using different tiny MLPs, based on importance scores of Gaussians as fitting weights. We experiment on multiple datasets, achieving a 91-times average model size reduction without harming the visual quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_23162 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations Tang, Zhenyu Feng, Chaoran Cheng, Xinhua Yu, Wangbo Zhang, Junwu Liu, Yuan Long, Xiaoxiao Wang, Wenping Yuan, Li Computer Vision and Pattern Recognition 3D Gaussian Splatting (3DGS) achieves impressive quality and rendering speed, but with millions of 3D Gaussians and significant storage and transmission costs. In this paper, we aim to develop a simple yet effective method called NeuralGS that compresses the original 3DGS into a compact representation. Our observation is that neural fields like NeRF can represent complex 3D scenes with Multi-Layer Perceptron (MLP) neural networks using only a few megabytes. Thus, NeuralGS effectively adopts the neural field representation to encode the attributes of 3D Gaussians with MLPs, only requiring a small storage size even for a large-scale scene. To achieve this, we adopt a clustering strategy and fit the Gaussians within each cluster using different tiny MLPs, based on importance scores of Gaussians as fitting weights. We experiment on multiple datasets, achieving a 91-times average model size reduction without harming the visual quality. |
| title | NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.23162 |