GaussFusion: 通过几何引导的视频生成提升野外3D重建
GaussFusion: Improving 3D Reconstruction in the Wild with A Geometry-Informed Video Generator
AI总结：
GaussFusion通过几何引导的视频生成技术改进3DGS重建，解决常见伪影问题，提供更鲁棒的3D重建方法，实现在新型视图合成中的最佳性能。
摘要：
我们提出了GaussFusion，一种通过几何引导的视频生成技术来改进3D高斯溅射（3DGS）在野外的重建方法。GaussFusion缓解了常见的3DGS伪影，包括由相机姿态误差、不完全覆盖和噪声几何初始化引起的漂浮物、闪烁和模糊。与以往仅限于单一重建流程的RGB方法不同，我们的方法引入了几何引导的视频到视频生成器，该生成器在基于优化和前馈的方法上都优化3DGS渲染。给定现有重建，我们渲染一个高斯原语视频缓冲区，编码深度、法线、不透明度和协方差，该生成器将其细化为时间上一致且无伪影的帧。我们进一步引入了伪影合成流程，模拟多样的退化模式，确保鲁棒性和泛化能力。GaussFusion在新型视图合成基准上实现了最先进的性能，且一个高效的变种在15 FPS下运行，保持相似性能，使交互式3D应用成为可能。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
