Robust 3D Gaussian Splatting for Novel View Synthesis in Presence of Distractors

Paul Ungermann, Armin Ettenhofer, Matthias Nießner, Barbara Roessle
Technical University of Munich

Due to distractors in the scene 3D Gaussian Splatting creates floating artifacts in the image (highlighted with circles). Our method mitigates artifacts due to violations of the static scene assumption for Gaussian Splatting. As a key element to our approach, we optimize for semantic distractor masks simultaneous to the scene optimization, which allow us to effectively ignore distractors.

Abstract

3D Gaussian Splatting has shown impressive novel view synthesis results; nonetheless, it is vulnerable to dynamic objects polluting the input data of an otherwise static scene, so called distractors. Distractors have severe impact on the rendering quality as they get represented as view-dependent effects or result in floating artifacts. Our goal is to identify and ignore such distractors during the 3D Gaussian optimization to obtain a clean reconstruction. To this end, we take a self-supervised approach that looks at the image residuals during the optimization to determine areas that have likely been falsified by a distractor. In addition, we leverage a pretrained segmentation network to provide object awareness, enabling more accurate exclusion of distractors. This way, we obtain segmentation masks of distractors to effectively ignore them in the loss formulation. We demonstrate that our approach is robust to various distractors and significantly improves rendering quality on distractor-polluted scenes, improving PSNR by 1.86dB compared to 3D Gaussian Splatting.

Pipeline

Robust 3D Gaussian Splatting computes semantic distractor masks to ignore distractors during optimization. Furthermore, it leverages SegmentAnything to provide object awareness for more accurate exclusion of distractors.

Video

Robust Gaussian Splatting (ours)

Gaussian Splatting

BibTeX

@article{ungermann2024robustgaussians,
  author    = {Ungermann, Paul and Ettenhofer, Armin and Nie{\ss}ner, Matthias and Roessle, Barbara},
  title     = {Robust 3D Gaussian Splatting for Novel ViewSynthesis in Presence of Distractors},
  journal   = {GCPR},
  year      = {2024}
}