Abstract

Recently, Gaussian Splatting has sparked a new trend in the field of computer vision. Apart from novel view synthesis, it has also been extended to the area of multi-view reconstruction. The latest methods facilitate complete, detailed surface reconstruction while ensuring fast training speed. However, these methods still require dense input views, and their output quality significantly degrades with sparse views. We observed that the Gaussian primitives tend to overfit the few training views, leading to noisy floaters and incomplete reconstruction surfaces. In this paper, we present an innovative sparse-view reconstruction framework that leverages intra-view depth and multi-view feature consistency to achieve remarkably accurate surface reconstruction. Specifically, we utilize monocular depth ranking information to supervise the consistency of depth distribution within patches and employ a smoothness loss to enhance the continuity of the distribution. To achieve finer surface reconstruction, we optimize the absolute position of depth through multi-view projection features. Extensive experiments on DTU and BlendedMVS demonstrate that our method outperforms state-of-the-art methods with a speedup of 60x to 200x, achieving swift and fine-grained mesh reconstruction without the need for costly pre-training.

Method

overview

Overview of FatesGS. Starting with a set of sparse input views, we initialize 2D Gaussians using COLMAP and employ splatting to render RGB images and depth maps. To enhance the geometric learning process, we integrate ranking information from monocular depth estimation and apply depth smoothing to ensure intra-view depth consistency. To further refine the geometry, we align the multi-view features extracted by projecting estimated surface points onto the source images.

Comparison Results

Comparison on DTU Dataset

Qualitative comparison of reconstruction results on DTU dataset with different sparse settings.

overview

Comparison on BlendedMVS Dataset

Qualitative comparison of reconstruction results on BlendedMVS dataset with three input views.

overview

Ablation Study

overview

Visual comparison of ablation study on DTU Scan 83. The transition of the error maps from blue to yellow indicates larger reconstruction errors.

BibTeX 

@inproceedings{huang2025fatesgs,
    title={FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency},
    author={Han Huang and Yulun Wu and Chao Deng and Ge Gao and Ming Gu and Yu-Shen Liu},
    booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
    year={2025}
}

Acknowledgements

The framework is built upon 2DGS and 3DGS. Thanks to the authors for their great work.
The website template was borrowed from Ref-NeRF and Michaël Gharbi.