UNG-GS: Uncertainty-Aware Normal-Guided Gaussian Splatting for Surface Reconstruction from Sparse Image Sequences

Zhen Tan, Xieyuanli Chen, Jinpu Zhang, Lei Feng, Dewen Hu*
| arXiv | Data |

---

📢 Overview

We propose Uncertainty-aware Normal-Guided Gaussian Splatting (UNG-GS), a novel framework that introduces an explicit Spatial Uncertainty Field (SUF) to quantify geometric uncertainty within the 3DGS pipeline.

---

📊 Visualization

Figure 1: Multi-scene reconstruction and rendering results.

Figure 2: Visualization of the spatial uncertainty field.

---

📚 Citation

If you find our work or code useful, please consider citing:

@article{tan2025uncertainty,
  title={Uncertainty-Aware Normal-Guided Gaussian Splatting for Surface Reconstruction from Sparse Image Sequences},
  author={Tan, Zhen and Chen, Xieyuanli and Zhang, Jinpu and Feng, Lei and Hu, Dewen},
  journal={arXiv preprint arXiv:2503.11172},
  year={2025}
}