AI RESEARCH

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

arXiv CS.CV

ArXi:2603.25053v1 Announce Type: new 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