MGS-SLAM：基于深度平滑正则化的单目稀疏跟踪与高斯建图
MGS-SLAM: Monocular Sparse Tracking and Gaussian Mapping with Depth Smooth Regularization
AI总结：
该研究提出MGS-SLAM，通过联合优化稀疏视觉里程计与3D Gaussian Splatting，并结合MVS深度监督、深度平滑损失和SDAR，提升单目VSLAM的位姿估计、新视角合成与几何重建精度。
摘要：
本文提出了一种基于高斯泼溅（Gaussian Splatting）的稠密视觉同时定位与建图（VSLAM）新框架。近年来，基于高斯泼溅的SLAM已展现出良好效果。然而，在单目场景中，重建的高斯地图缺乏几何精度，并且跟踪能力较弱。为解决这些局限，我们首次将稀疏视觉里程计跟踪与3D Gaussian Splatting场景表示进行联合优化。我们使用快速多视图立体（MVS）网络在视觉里程计关键帧窗口上获取深度图，用于对高斯地图进行几何监督。此外，我们提出了深度平滑损失和稀疏-稠密调整环（SDAR），以降低估计深度图的负面影响，并保持视觉里程计与高斯地图之间的尺度一致性。我们在多个合成数据集和真实世界数据集上评估了所提系统。我们的位姿估计精度超过现有方法，达到当前最优水平。此外，在新视角合成和几何重建保真度方面，该方法也优于以往的单目方法。
英文摘要：
This letter introduces a novel framework for dense Visual Simultaneous Localization and Mapping (VSLAM) based on Gaussian Splatting. Recently, SLAM based on Gaussian Splatting has shown promising results. However, in monocular scenarios, the Gaussian maps reconstructed lack geometric accuracy and exhibit weaker tracking capability. To address these limitations, we jointly optimize sparse visual odometry tracking and 3D Gaussian Splatting scene representation for the first time. We obtain depth maps on visual odometry keyframe windows using a fast Multi-View Stereo (MVS) network for the geometric supervision of Gaussian maps. Furthermore, we propose a depth smooth loss and Sparse-Dense Adjustment Ring (SDAR) to reduce the negative effect of estimated depth maps and preserve the consistency in scale between the visual odometry and Gaussian maps. We have evaluated our system across various synthetic and real-world datasets. The accuracy of our pose estimation surpasses existing methods and achieves state-of-the-art. Additionally, it outperforms previous monocular methods in terms of novel view synthesis and geometric reconstruction fidelities.
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。
