Sources of Uncertainty in 3D Scene Reconstruction
This work addresses uncertainty quantification for 3D reconstruction methods, which is incremental as it builds on existing NeRF and GS techniques.
The paper tackles the problem of uncertainty in 3D scene reconstruction by introducing a taxonomy for uncertainty sources and extending NeRF and 3D Gaussian Splatting methods with uncertainty estimation techniques, resulting in an empirical study that assesses their sensitivity.
The process of 3D scene reconstruction can be affected by numerous uncertainty sources in real-world scenes. While Neural Radiance Fields (NeRFs) and 3D Gaussian Splatting (GS) achieve high-fidelity rendering, they lack built-in mechanisms to directly address or quantify uncertainties arising from the presence of noise, occlusions, confounding outliers, and imprecise camera pose inputs. In this paper, we introduce a taxonomy that categorizes different sources of uncertainty inherent in these methods. Moreover, we extend NeRF- and GS-based methods with uncertainty estimation techniques, including learning uncertainty outputs and ensembles, and perform an empirical study to assess their ability to capture the sensitivity of the reconstruction. Our study highlights the need for addressing various uncertainty aspects when designing NeRF/GS-based methods for uncertainty-aware 3D reconstruction.