GaussFusion: Improving 3D Reconstruction in the Wild with A Geometry-Informed Video Generator

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Liyuan, Narayana, Manjunath, Stary, Michal, Hutchcroft, Will, Wetzstein, Gordon, Armeni, Iro
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911549561503744
author Zhu, Liyuan
Narayana, Manjunath
Stary, Michal
Hutchcroft, Will
Wetzstein, Gordon
Armeni, Iro
author_facet Zhu, Liyuan
Narayana, Manjunath
Stary, Michal
Hutchcroft, Will
Wetzstein, Gordon
Armeni, Iro
contents We present GaussFusion, a novel approach for improving 3D Gaussian splatting (3DGS) reconstructions in the wild through geometry-informed video generation. GaussFusion mitigates common 3DGS artifacts, including floaters, flickering, and blur caused by camera pose errors, incomplete coverage, and noisy geometry initialization. Unlike prior RGB-based approaches limited to a single reconstruction pipeline, our method introduces a geometry-informed video-to-video generator that refines 3DGS renderings across both optimization-based and feed-forward methods. Given an existing reconstruction, we render a Gaussian primitive video buffer encoding depth, normals, opacity, and covariance, which the generator refines to produce temporally coherent, artifact-free frames. We further introduce an artifact synthesis pipeline that simulates diverse degradation patterns, ensuring robustness and generalization. GaussFusion achieves state-of-the-art performance on novel-view synthesis benchmarks, and an efficient variant runs in real time at 15 FPS while maintaining similar performance, enabling interactive 3D applications.
format Preprint
id arxiv_https___arxiv_org_abs_2603_25053
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle GaussFusion: Improving 3D Reconstruction in the Wild with A Geometry-Informed Video Generator
Zhu, Liyuan
Narayana, Manjunath
Stary, Michal
Hutchcroft, Will
Wetzstein, Gordon
Armeni, Iro
Computer Vision and Pattern Recognition
We present GaussFusion, a novel approach for improving 3D Gaussian splatting (3DGS) reconstructions in the wild through geometry-informed video generation. GaussFusion mitigates common 3DGS artifacts, including floaters, flickering, and blur caused by camera pose errors, incomplete coverage, and noisy geometry initialization. Unlike prior RGB-based approaches limited to a single reconstruction pipeline, our method introduces a geometry-informed video-to-video generator that refines 3DGS renderings across both optimization-based and feed-forward methods. Given an existing reconstruction, we render a Gaussian primitive video buffer encoding depth, normals, opacity, and covariance, which the generator refines to produce temporally coherent, artifact-free frames. We further introduce an artifact synthesis pipeline that simulates diverse degradation patterns, ensuring robustness and generalization. GaussFusion achieves state-of-the-art performance on novel-view synthesis benchmarks, and an efficient variant runs in real time at 15 FPS while maintaining similar performance, enabling interactive 3D applications.
title GaussFusion: Improving 3D Reconstruction in the Wild with A Geometry-Informed Video Generator
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.25053