iGaussian：基于前馈三维高斯泼溅反演的实时相机位姿估计
iGaussian: Real-Time Camera Pose Estimation via Feed-Forward 3D Gaussian Splatting Inversion
AI总结：
针对现有基于3D高斯的相机位姿估计方法迭代优化开销大、难以实时运行的问题，提出两阶段前馈框架iGaussian，通过直接高斯反演实现位姿估计，在多数据集上精度更高，移动机器人上速度较优化类方法提升10倍。
摘要：
近期同步定位与地图构建（SLAM）及视觉导航领域的发展趋势已将3D Gaussians作为首选场景表示方式，这凸显了利用预构建高斯模型从单张图像估计相机位姿的重要性。然而现有方法通常依赖迭代式“渲染-比较-优化”循环：首先使用NeRF或Gaussian Splatting渲染候选视图，再将其与目标图像对比，最后利用差异更新位姿。这种多轮过程会产生显著计算开销，阻碍机器人场景下的实时性能表现。本文提出iGaussian，这是一种两阶段前馈框架，可通过直接3D高斯反演实现实时相机位姿估计。该方法首先借助搭载空间均匀采样与引导注意力机制、基于高斯场景先验的位姿回归网络，回归得到粗略的6自由度（6DoF）位姿，再通过特征匹配与多模型融合对其进行优化。核心贡献在于我们设计的互相关模块，该模块无需可微渲染即可对齐图像嵌入与3D高斯属性，同时搭配加权多视图预测器，融合从多个策略性采样视点获取的特征。在NeRF Synthetic、Mip-NeRF 360以及T&T+DB数据集上的实验结果表明，该方法相较此前方法性能提升显著，中位旋转误差降至0.2°，同时在移动机器人上实现了2.87 FPS的跟踪速度，相较基于优化的方法速度提升达10倍。代码地址：https://github.com/pythongod-exe/iGaussian
英文摘要：
Recent trends in SLAM and visual navigation have embraced 3D Gaussians as the preferred scene representation, highlighting the importance of estimating camera poses from a single image using a pre-built Gaussian model. However, existing approaches typically rely on an iterative \textit{render-compare-refine} loop, where candidate views are first rendered using NeRF or Gaussian Splatting, then compared against the target image, and finally, discrepancies are used to update the pose. This multi-round process incurs significant computational overhead, hindering real-time performance in robotics. In this paper, we propose iGaussian, a two-stage feed-forward framework that achieves real-time camera pose estimation through direct 3D Gaussian inversion. Our method first regresses a coarse 6DoF pose using a Gaussian Scene Prior-based Pose Regression Network with spatial uniform sampling and guided attention mechanisms, then refines it through feature matching and multi-model fusion. The key contribution lies in our cross-correlation module that aligns image embeddings with 3D Gaussian attributes without differentiable rendering, coupled with a Weighted Multiview Predictor that fuses features from Multiple strategically sampled viewpoints. Experimental results on the NeRF Synthetic, Mip-NeRF 360, and T\&T+DB datasets demonstrate a significant performance improvement over previous methods, reducing median rotation errors to 0.2° while achieving 2.87 FPS tracking on mobile robots, which is an impressive 10 times speedup compared to optimization-based approaches. Code: https://github.com/pythongod-exe/iGaussian
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。
