Research map / 3D vision / Gaussian splatting
NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations
Zhenyu Tang, Chaoran Feng, Xinhua Cheng, Wangbo Yu, Junwu Zhang, Yuan Liu, Xiao-Xiao Long, Wenping Wang, Li Yuan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add attribute-based clustering with separate tiny MLPs per cluster (ablation) | NeuralGS · Deep Blending | PSNR: 23.71 -> 28.98 | Improved |
| add attribute-based clustering with separate tiny MLPs per cluster (ablation) | NeuralGS · Deep Blending | model size: 2.69 MB -> 5.79 MB | Hurt |
| add importance weighting to fitting loss (ablation) | NeuralGS · Deep Blending | PSNR: 28.98 -> 29.67 | Improved |
| add importance weighting to fitting loss (ablation) | NeuralGS · Deep Blending | LPIPS: 0.283 -> 0.264 | Improved |
| add frequency loss during fine-tuning (ablation) | NeuralGS · Deep Blending | PSNR: 29.67 -> 30.09 | Improved |
| add frequency loss during fine-tuning (ablation) | NeuralGS · Deep Blending | model size: 5.75 MB -> 5.76 MB | No clear effect |
| NeuralGS vs original 3DGS compression | NeuralGS · Deep Blending | model size: 672.8 MB -> 5.76 MB | Improved |
| NeuralGS vs 3DGS rendering speed | NeuralGS · Mip-NeRF360 | FPS: 112 -> 205 | Improved |
| NeuralGS vs HAC rendering speed | NeuralGS · Deep Blending | FPS: 159 -> 217 | Improved |
| use NeuralGS instead of HAC | NeuralGS · Deep Blending | model size: 4.62 MB -> 5.76 MB | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CompGS — CompGS: Smaller and Faster Gaussian Splatting with Vector Quantization (ECCV 2024) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Compact 3D Gaussian Representation for Radiance Field (CVPR 2024 · similarity 0.93)
- ContextGS : Compact 3D Gaussian Splatting with Anchor Level Context Model (NeurIPS 2024 · similarity 0.92)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.91)
- 3D Student Splatting and Scooping (CVPR 2025 · similarity 0.91)
- Gaussian Herding across Pens: An Optimal Transport Perspective on Global Gaussian Reduction for 3DGS (NeurIPS 2025 · similarity 0.90)
- Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields (CVPR 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
