3DGS SLAM

3DGS: 3D Gaussian Splatting · SLAM: Simultaneous Localization and Mapping

The field studies how a camera can track its motion while incrementally building a detailed, view-renderable 3D map from images and other sensors. Systems optimize Gaussian maps and camera poses using photometric and geometric cues, with approaches spanning monocular, RGB-D, visual-inertial, and LiDAR setups; some add learned depth or pose priors, loop closure, semantic features, or uncertainty estimates. Work also extends Gaussian maps to dynamic and deformable scenes, and commonly evaluates tracking and rendering on benchmarks such as TUM RGB-D, Replica, ScanNet, and KITTI. A central tension is balancing map quality and robust global tracking against optimization time, memory use, and reliability in challenging conditions such as motion, blur, and large outdoor scenes.

Latest papers

63 papers6 papers in the last four weeks, down 45% on the four weeks before. 0.0% of all new papers.

Robust 3DGS-based SLAM via Adaptive Kernel Smoothing
Nov 28, 2025Shouhe Zhang, Dayong Ren, Wen Jie Li +4Simultaneous Localization and Mapping 3D Gaussian Splatting
School of Computer Science and Technology, Xinjiang University · The National Key Laboratory for Novel Software Technology, Nanjing University · Southwest University of Political Science and Law +2
LEGO-SLAM: Language-Embedded Gaussian Optimization SLAM
Nov 20, 2025Sibaek Lee, Seongbo Ha, Kyeongsu Kang +33D Gaussian Splatting 3DGS SLAM
Department of Intelligent Robotics, Sungkyunkwan University, Suwon, South Korea
DINO-SLAM: DINO-informed RGB-D SLAM for Neural Implicit and Explicit Representations
Jul 25, 2025Ziren Gong, Xiaohan Li, Fabio Tosi +4Simultaneous Localization and Mapping Neural Representation Geometry
University of Bologna, Italy · Faculty of Dentistry, The University of Hong Kong, China · Rawmantic AI, China +1

Nov 28, 2025J/K move · Enter open · S save

Robust 3DGS-based SLAM via Adaptive Kernel Smoothing

Shouhe Zhang, Dayong Ren, Wen Jie Li, Piaopiao Yu+3

School of Computer Science and Technology, Xinjiang University · The National Key Laboratory for Novel Software Technology, Nanjing University · Southwest University of Political Science and Law+2

Simultaneous Localization and Mapping 3D Gaussian Splatting 3DGS SLAM Camera Pose Estimation

Open paper AskPDF arXiv

In this paper, we challenge the conventional notion in 3DGS-SLAM that rendering quality is the primary determinant of tracking accuracy. We argue that, compared to solely pursuing a perfect scene representation, it is more critical to enhance the robustness of the rasterization process against parameter errors to ensure stable camera pose tracking. To address this challenge, we propose a novel approach that leverages a smooth kernel strategy to enhance the robustness of 3DGS-based SLAM. Unlike conventional methods that focus solely on minimizing rendering error, our core insight is to make the rasterization process more resilient to imperfections in the 3DGS parameters. We hypothesize that by allowing each Gaussian to influence a smoother, wider distribution of pixels during rendering, we can mitigate the detrimental effects of parameter noise from outlier Gaussians. This approach intentionally introduces a controlled blur to the rendered image, which acts as a regularization term, stabilizing the subsequent pose optimization. While a complete redesign of the rasterization pipeline is an ideal solution, we propose a practical and effective alternative that is readily integrated into existing 3DGS frameworks. Our method, termed Corrective Blurry KNN (CB-KNN), adaptively modifies the RGB values and locations of the K-nearest neighboring Gaussians within a local region. This dynamic adjustment generates a smoother local rendering, reducing the impact of erroneous GS parameters on the overall image. Experimental results demonstrate that our approach, while maintaining the overall quality of the scene reconstruction (mapping), significantly improves the robustness and accuracy of camera pose tracking.

Latest papers

Robust 3DGS-based SLAM via Adaptive Kernel Smoothing

LEGO-SLAM: Language-Embedded Gaussian Optimization SLAM

DINO-SLAM: DINO-informed RGB-D SLAM for Neural Implicit and Explicit Representations