概述。我们提出了GSplatLoc,这是一种新颖的相机定位方法,它利用3D高斯溅射的可微渲染能力进行高效且准确的姿态估计。通过将姿态估计表述为完全可微框架内的基于梯度的优化问题,GSplatLoc能够使用从预先存在的3D高斯场景表示中渲染的深度信息直接优化相机姿态。这种方法使我们能够实现适合实时应用的高精度定位。 动机。使用...
GSplatLoc: Grounding Keypoint Descriptors into 3D Gaussian Splatting for Improved Visual Localization [Paper] [Project Page] [Video] Authors: Gennady Sidorov, Malik Mohrat, Ksenia Lebedeva, Ruslan Rakhimov, Sergey Kolyubin This repository contains the code for the paper "GSplatLoc: Grounding Key...
To address these challenges, we utilize recent advancements in novel view synthesis, particularly 3D Gaussian Splatting (3DGS), to enhance localization. 3DGS allows for the compact encoding of both 3D geometry and scene appearance with its spatial features. Our method leverages the dense description...