LoGS: Visual Localization via Gaussian Splatting with Fewer Training Images
Not on the map. Why?
Abstract (arXiv)
Visual localization involves estimating a query image's 6-DoF (degrees of freedom) camera pose, which is a fundamental component in various computer vision and robotic tasks. This paper presents LoGS, a vision-based localization pipeline utilizing the 3D Gaussian Splatting (GS) technique as scene representation. This novel representation allows high-quality novel view synthesis. During the mapping phase, structure-from-motion (SfM) is applied first, followed by the generation of a GS map. During localization, the initial position is obtained through image retrieval, local feature matching coupled with a PnP solver, and then a high-precision pose is achieved through the analysis-by-synthesis manner on the GS map. Experimental results on four large-scale datasets demonstrate the proposed approach's SoTA accuracy in estimating camera poses and robustness under challenging few-shot conditions.
Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 37 of its 46 references that are in the map. First posted on arXiv 15 Oct 2024.
Builds on
37 of 46 references.
References · 37
- 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
- 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,032
- 2021Vision Transformers for Dense Predictioncited by 1,803
- 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,443
- 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,320
- 2015NetVLAD: CNN architecture for weakly supervised place recognitioncited by 941
- 2019Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelinescited by 672
- 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 559
- 2016Direct Sparse Odometrycited by 485
- 2023Gaussian Splatting SLAMcited by 415
- 2020INeRF: Inverting Neural Radiance Fields for Pose Estimationcited by 382
- 2023FSGS: Real-Time Few-shot View Synthesis using Gaussian Splattingcited by 279
25 more
- 2016DSAC - Differentiable RANSAC for Camera Localizationcited by 266
- 2017Geometric Loss Functions for Camera Pose Regression with Deep Learningcited by 224
- 2024A Survey on 3D Gaussian Splattingcited by 224
- 2019Understanding the Limitations of CNN-based Absolute Camera Pose Regressioncited by 177
- 2020Visual Camera Re-Localization from RGB and RGB-D Images Using DSACcited by 176
- 2021Learning Multi-Scene Absolute Pose Regression with Transformerscited by 108
- 2023Accelerated Coordinate Encoding: Learning to Relocalize in Minutes using RGB and Posescited by 107
- 2019Hierarchical Scene Coordinate Classification and Regression for Visual Localizationcited by 93
- 2022DFNet: Enhance Absolute Pose Regression with Direct Feature Matchingcited by 82
- 2021On the Limits of Pseudo Ground Truth in Visual Camera Re-localisationcited by 63
- 2016Learning to Navigate the Energy Landscapecited by 62
- 2021Direct-PoseNet: Absolute Pose Regression with Photometric Consistencycited by 61
- 2017VidLoc: A Deep Spatio-Temporal Model for 6-DoF Video-Clip Relocalizationcited by 58
- 2022Parallel Inversion of Neural Radiance Fields for Robust Pose Estimationcited by 55
- 2024GLACE: Global Local Accelerated Coordinate Encodingcited by 48
- 2024Map-Relative Pose Regression for Visual Re-Localizationcited by 42
- 2023Neural Refinement for Absolute Pose Regression with Feature Synthesiscited by 37
- 2022Visual Localization via Few-Shot Scene Region Classificationcited by 36
- 2024The NeRFect Match: Exploring NeRF Features for Visual Localizationcited by 32
- 2018Leveraging Deep Visual Descriptors for Hierarchical Efficient Localizationcited by 29
- 2021VS-Net: Voting with Segmentation for Visual Localizationcited by 29
- 20246DGS: 6D Pose Estimation from a Single Image and a 3D Gaussian Splatting Modelcited by 28
- 2024The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinementcited by 21
- 2024HR-APR: APR-agnostic Framework with Uncertainty Estimation and Hierarchical Refinement for Camera Relocalisationcited by 16
- 2020Curved patch mapping and tracking for irregular terrain modeling: Application to bipedal robot foot placementcited by 3
Built on by
Citing papers per year: 2025: 2; 2026: 3.
5 papers.
- 2025From Sparse to Dense: Camera Relocalization with Scene-Specific Detector from Feature Gaussian Splattingcited by 14
- 2025Hi^2-GSLoc: Dual-Hierarchical Gaussian-Specific Visual Relocalization for Remote Sensingcited by 0
- 2026Hierarchical Visual Relocalization with Nearest View Synthesis from Feature Gaussian Splattingcited by 0
- 2026LSGS-Loc: Towards Robust 3DGS-Based Visual Localization for Large-Scale UAV Scenarioscited by 0
- 2026MLP Splatting: Object-Centric Neural Fieldscited by 0
- 2026MLP Splatting: Object-Centric Neural Fields2 Jun 2026
- 2026LSGS-Loc: Towards Robust 3DGS-Based Visual Localization for Large-Scale UAV Scenarios7 Apr 2026
- 2026Hierarchical Visual Relocalization with Nearest View Synthesis from Feature Gaussian Splatting31 Mar 2026
- 2025Hi^2-GSLoc: Dual-Hierarchical Gaussian-Specific Visual Relocalization for Remote Sensing21 Jul 2025
- 2025From Sparse to Dense: Camera Relocalization with Scene-Specific Detector from Feature Gaussian Splatting25 Mar 2025