mlsanityData to 7 Oct 2026

VarSplat: Uncertainty-aware 3D Gaussian Splatting for Robust RGB-D SLAM

arXiv 2603.09673 ↗First posted 10 Mar 2026cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Simultaneous Localization and Mapping (SLAM) with 3D Gaussian Splatting (3DGS) enables fast, differentiable rendering and high-fidelity reconstruction across diverse real-world scenes. However, existing 3DGS-SLAM approaches handle measurement reliability implicitly, making pose estimation and global alignment susceptible to drift in low-texture regions, transparent surfaces, or areas with complex reflectance properties. To this end, we introduce VarSplat, an uncertainty-aware 3DGS-SLAM system that explicitly learns per-splat appearance variance. By using the law of total variance with alpha compositing, we then render differentiable per-pixel uncertainty map via efficient, single-pass rasterization. This map guides tracking, submap registration, and loop detection toward focusing on reliable regions and contributes to more stable optimization. Experimental results on Replica (synthetic) and TUM-RGBD, ScanNet, and ScanNet++ (real-world) show that VarSplat improves robustness and achieves competitive or superior tracking, mapping, and novel view synthesis rendering compared to existing studies for dense RGB-D SLAM.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 40 of its 56 references that are in the map. First posted on arXiv 10 Mar 2026.

Builds on

40 of 56 references.

References · 40

  1. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,141
  2. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  3. 2015NetVLAD: CNN architecture for weakly supervised place recognitioncited by 935
  4. 2019The Replica Dataset: A Digital Replica of Indoor Spacescited by 885
  5. 2020ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAMcited by 862
  6. 2023ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenescited by 747
  7. 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 586
  8. 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 556
  9. 2022Point-NeRF: Point-based Neural Radiance Fieldscited by 503
  10. 2021iMAP: Implicit Mapping and Positioning in Real-Timecited by 476
  11. 2023Gaussian Splatting SLAMcited by 410
  12. 2023SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAMcited by 386
28 more
  1. 2023GS-SLAM: Dense Visual SLAM with 3D Gaussian Splattingcited by 278
  2. 2023Co-SLAM: Joint Coordinate and Sparse Parametric Encodings for Neural Real-Time SLAMcited by 183
  3. 2023Point-SLAM: Dense Neural Point Cloud-based SLAMcited by 160
  4. 2022ESLAM: Efficient Dense SLAM System Based on Hybrid Representation of Signed Distance Fieldscited by 158
  5. 2023Gaussian-SLAM: Photo-realistic Dense SLAM with Gaussian Splattingcited by 152
  6. 2024WildGaussians: 3D Gaussian Splatting in the Wildcited by 150
  7. 2022Vox-Fusion: Dense Tracking and Mapping with Voxel-based Neural Implicit Representationcited by 149
  8. 2023Photo-SLAM: Real-time Simultaneous Localization and Photorealistic Mapping for Monocular, Stereo, and RGB-D Camerascited by 139
  9. 2023GO-SLAM: Global Optimization for Consistent 3D Instant Reconstructioncited by 132
  10. 2020DeepFactors: Real-Time Probabilistic Dense Monocular SLAMcited by 117
  11. 2023Bayes' Rays: Uncertainty Quantification for Neural Radiance Fieldscited by 90
  12. 2024PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splattingcited by 85
  13. 2021Stochastic Neural Radiance Fields: Quantifying Uncertainty in Implicit 3D Representationscited by 80
  14. 2022Conditional-Flow NeRF: Accurate 3D Modelling with Reliable Uncertainty Quantificationcited by 71
  15. 2020DI-Fusion: Online Implicit 3D Reconstruction with Deep Priorscited by 70
  16. 2023FisherRF: Active View Selection and Uncertainty Quantification for Radiance Fields using Fisher Informationcited by 63
  17. 2024CG-SLAM: Efficient Dense RGB-D SLAM in a Consistent Uncertainty-aware 3D Gaussian Fieldcited by 61
  18. 2024LoopSplat: Loop Closure by Registering 3D Gaussian Splatscited by 59
  19. 2024Loopy-SLAM: Dense Neural SLAM with Loop Closurescited by 48
  20. 2025WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environmentscited by 48
  21. 2023MIPS-Fusion: Multi-Implicit-Submaps for Scalable and Robust Online Neural RGB-D Reconstructioncited by 24
  22. 2023UncLe-SLAM: Uncertainty Learning for Dense Neural SLAMcited by 19
  23. 2024Modeling uncertainty for Gaussian Splattingcited by 17
  24. 2024Modeling Uncertainty in 3D Gaussian Splatting through Continuous Semantic Splattingcited by 12
  25. 2024Bayesian NeRF: Quantifying Uncertainty with Volume Density for Neural Implicit Fieldscited by 9
  26. 2024GLC-SLAM: Gaussian Splatting SLAM with Efficient Loop Closurecited by 7
  27. 2024Uncertainty-aware Active Learning of NeRF-based Object Models for Robot Manipulators using Visual and Re-orientation Actionscited by 4
  28. 2024Uni-SLAM: Uncertainty-Aware Neural Implicit SLAM for Real-Time Dense Indoor Scene Reconstructioncited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026When Variance Is Not an Error Map: Calibrated Uncertainty for Radiative Gaussian Splatting in Sparse-View CTcited by 1
  2. 2026MoPe: Motion Permanence for Robust Monocular Gaussian Mapping in Dynamic Environmentscited by 0
  3. 2026View-Structured Conformal Prediction for 3D Gaussian Splattingcited by 0