无需训练即可在3D高斯点云中精准定位相机，支持新场景泛化。
GSVisLoc: Generalizable Visual Localization for Gaussian Splatting Scene Representations
- 利用3D高斯点云生成场景特征，与图像块特征匹配定位
- 三步流程实现粗匹配、精匹配与位姿优化，精度领先现有方法
- 直接适配已有3DGS模型，无需重训练或额外参考图
我们提出GSVisLoc，一种针对3D高斯点云（3DGS）场景表示的视觉定位方法。给定一个3DGS场景模型和查询图像，目标是估计相机的位置与姿态。通过鲁棒地匹配场景特征与图像特征实现：场景特征由下采样并编码3D高斯点生成，图像特征则来自图像块的编码。算法分三步进行：先进行粗匹配，再精细匹配，最后通过位姿优化获得精确估计。关键优势在于，本方法直接利用显式的3DGS场景表示进行定位，无需任何修改、重新训练或额外参考图像。我们在室内外场景上评估了该方法，在标准基准上表现优异，超越现有基于3DGS的基线方法。此外，该方法对未见过的新场景具有良好的泛化能力，无需额外训练。
原文摘要 · Abstract (English)
We introduce GSVisLoc, a visual localization method designed for 3D Gaussian Splatting (3DGS) scene representations. Given a 3DGS model of a scene and a query image, our goal is to estimate the camera's position and orientation. We accomplish this by robustly matching scene features to image features. Scene features are produced by downsampling and encoding the 3D Gaussians while image features are obtained by encoding image patches. Our algorithm proceeds in three steps, starting with coarse matching, then fine matching, and finally by applying pose refinement for an accurate final estimate. Importantly, our method leverages the explicit 3DGS scene representation for visual localization without requiring modifications, retraining, or additional reference images. We evaluate GSVisLoc on both indoor and outdoor scenes, demonstrating competitive localization performance on standard benchmarks while outperforming existing 3DGS-based baselines. Moreover, our approach generalizes effectively to novel scenes without additional training.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。