SpotLessSplats: Ignoring Distractors in 3D Gaussian Splatting

Sara Sabour, Lily Goli, Georgios Kopanas, Mark J. Matthews, Dmitry Lagun, Leonidas Guibas, Alec Jacobson, David J. Fleet, Andrea Tagliasacchi

Keywords: Robust Radiance Fields, 3D Gaussian Splatting, In the Wild Capture

Abstract:

3D Gaussian Splatting (3DGS) is a promising technique for 3D reconstruction, offering efficient training and rendering speeds, making it suitable for real-time applications. However, current methods require highly controlled environments—no moving people or wind-blown elements, and consistent lighting—to meet the inter-view consistency assumption of 3DGS. This makes reconstruction of real-world captures problematic. We present SpotLessSplats, an approach that leverages pre-trained and general-purpose features coupled with robust optimization to effectively ignore transient distractors.
Our method achieves state-of-the-art reconstruction quality both visually and quantitatively, on casual captures.

Submission Number: 8