MM3DGS SLAM：利用视觉、深度与惯性测量的多模态三维高斯泼溅SLAM
MM3DGS SLAM: Multi-modal 3D Gaussian Splatting for SLAM Using Vision, Depth, and Inertial Measurements
AI总结：
本文提出MM3DGS，首次将三维高斯泼溅与未标定相机图像和惯性测量结合用于SLAM，通过融合惯性预积分、深度估计和光度质量损失，实现3倍跟踪提升和5%渲染质量提升，并发布多模态数据集UT-MM。
摘要：
同时定位与建图对于位置跟踪和场景理解至关重要。基于三维高斯的建图表示能够利用多个已标定位姿的相机实现逼真重建和场景的实时渲染。我们首次证明，将三维高斯用于建图表示，并结合未标定位姿的相机图像和惯性测量，能够实现精确的SLAM。我们的方法MM3DGS解决了先前基于神经辐射场表示的局限性，实现了更快的渲染、尺度感知以及改进的轨迹跟踪。我们的框架支持基于关键帧的建图与跟踪，利用的损失函数融合了来自预积分惯性测量的相对位姿变换、深度估计以及光度渲染质量度量。我们还发布了一个多模态数据集UT-MM，该数据集采集自配备相机和惯性测量单元的移动机器人。在该数据集多个场景上的实验评估表明，与当前最先进的3DGS SLAM相比，MM3DGS在跟踪性能上实现了3倍提升，在光度渲染质量上实现了5%的提升，同时能够实时渲染高分辨率稠密三维地图。项目网页：https://vita-group.github.io/MM3DGS-SLAM
英文摘要：
Simultaneous localization and mapping is essential for position tracking and scene understanding. 3D Gaussian-based map representations enable photorealistic reconstruction and real-time rendering of scenes using multiple posed cameras. We show for the first time that using 3D Gaussians for map representation with unposed camera images and inertial measurements can enable accurate SLAM. Our method, MM3DGS, addresses the limitations of prior neural radiance field-based representations by enabling faster rendering, scale awareness, and improved trajectory tracking. Our framework enables keyframe-based mapping and tracking utilizing loss functions that incorporate relative pose transformations from pre-integrated inertial measurements, depth estimates, and measures of photometric rendering quality. We also release a multi-modal dataset, UT-MM, collected from a mobile robot equipped with a camera and an inertial measurement unit. Experimental evaluation on several scenes from the dataset shows that MM3DGS achieves 3x improvement in tracking and 5% improvement in photometric rendering quality compared to the current 3DGS SLAM state-of-the-art, while allowing real-time rendering of a high-resolution dense 3D map. Project Webpage: https://vita-group.github.io/MM3DGS-SLAM
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。
