mlsanityData to 7 Oct 2026

iComMa: Inverting 3D Gaussian Splatting for Camera Pose Estimation via Comparing and Matching

arXiv 2312.09031 ↗First posted 14 Dec 2023cs.CVcited by 12

Not on the map. Why?

Abstract (arXiv)

We present a method named iComMa to address the 6D camera pose estimation problem in computer vision. Conventional pose estimation methods typically rely on the target's CAD model or necessitate specific network training tailored to particular object classes. Some existing methods have achieved promising results in mesh-free object and scene pose estimation by inverting the Neural Radiance Fields (NeRF). However, they still struggle with adverse initializations such as large rotations and translations. To address this issue, we propose an efficient method for accurate camera pose estimation by inverting 3D Gaussian Splatting (3DGS). Specifically, a gradient-based differentiable framework optimizes camera pose by minimizing the residual between the query image and the rendered image, requiring no training. An end-to-end matching module is designed to enhance the model's robustness against adverse initializations, while minimizing pixel-level comparing loss aids in precise pose estimation. Experimental results on synthetic and complex real-world data demonstrate the effectiveness of the proposed approach in challenging conditions and the accuracy of camera pose estimation.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 36 of its 42 references that are in the map. First posted on arXiv 14 Dec 2023.

Builds on

36 of 42 references.

References · 36

  1. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,141
  2. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,013
  3. 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,607
  4. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  5. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  6. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 768
  7. 2021BARF: Bundle-Adjusting Neural Radiance Fieldscited by 641
  8. 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 586
  9. 2023LightGlue: Local Feature Matching at Light Speedcited by 583
  10. 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 556
  11. 2022Point-NeRF: Point-based Neural Radiance Fieldscited by 503
  12. 2021iMAP: Implicit Mapping and Positioning in Real-Timecited by 476
24 more
  1. 2018PVNet: Pixel-wise Voting Network for 6DoF Pose Estimationcited by 385
  2. 2020INeRF: Inverting Neural Radiance Fields for Pose Estimationcited by 381
  3. 2020CosyPose: Consistent multi-view multi-object 6D pose estimationcited by 287
  4. 2018DeepIM: Deep Iterative Matching for 6D Pose Estimationcited by 283
  5. 2022NoPe-NeRF: Optimising Neural Radiance Field with No Pose Priorcited by 235
  6. 2017Real-Time Seamless Single Shot 6D Object Pose Predictioncited by 210
  7. 2022MegaPose: 6D Pose Estimation of Novel Objects via Render & Comparecited by 198
  8. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 182
  9. 2022ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformercited by 170
  10. 2020Objectron: A Large Scale Dataset of Object-Centric Videos in the Wild with Pose Annotationscited by 155
  11. 2022OnePose: One-Shot Object Pose Estimation without CAD Modelscited by 155
  12. 2022MatchFormer: Interleaving Attention in Transformers for Feature Matchingcited by 126
  13. 2023OnePose++: Keypoint-Free One-Shot Object Pose Estimation without CAD Modelscited by 114
  14. 2020Category Level Object Pose Estimation via Neural Analysis-by-Synthesiscited by 100
  15. 2022OSOP: A Multi-Stage One Shot Object Pose Estimation Frameworkcited by 81
  16. 2022NeRF-Pose: A First-Reconstruct-Then-Regress Approach for Weakly-supervised 6D Object Pose Estimationcited by 57
  17. 2022Zero-Shot Category-Level Object Pose Estimationcited by 57
  18. 2021Single-Stage Keypoint-Based Category-Level Object Pose Estimation from an RGB Imagecited by 42
  19. 2023PATS: Patch Area Transportation with Subdivision for Local Feature Matchingcited by 33
  20. 2023POPE: 6-DoF Promptable Pose Estimation of Any Object, in Any Scene, with One Referencecited by 30
  21. 2022Neural Correspondence Field for Object Pose Estimationcited by 21
  22. 2022Keypoint-Based Category-Level Object Pose Tracking from an RGB Sequence with Uncertainty Estimationcited by 19
  23. 2023PoseMatcher: One-shot 6D Object Pose Estimation by Deep Feature Matchingcited by 12
  24. 2022Nerfels: Renderable Neural Codes for Improved Camera Pose Estimationcited by 9

Built on by

Cited by year

Citing papers per year: 2024: 3; 2025: 9.

12 papers.

  1. 20243D Gaussian Splatting: Survey, Technologies, Challenges, and Opportunitiescited by 77
  2. 2025What Is The Best 3D Scene Representation for Robotics? From Geometric to Foundation Modelscited by 29
  3. 2024DDGS-CT: Direction-Disentangled Gaussian Splatting for Realistic Volume Renderingcited by 22
  4. 2025A Guide to Structureless Visual Localizationcited by 9
  5. 2025Digital Twin Generation from Visual Data: A Surveycited by 7
  6. 2025GS-LTS: 3D Gaussian Splatting-Based Adaptive Modeling for Long-Term Service Robotscited by 6
  7. 20253DGS_LSR:Large_Scale Relocation for Autonomous Driving Based on 3D Gaussian Splattingcited by 1
  8. 2025Camera Pose Refinement via 3D Gaussian Splattingcited by 1
  9. 2024Image Compression Using Novel View Synthesis Priorscited by 0
  10. 2025Improving Novel view synthesis of 360^circ Scenes in Extremely Sparse Views by Jointly Training Hemisphere Sampled Synthetic Imagescited by 0
  11. 2025SGLoc: Semantic Localization System for Camera Pose Estimation from 3D Gaussian Splatting Representationcited by 0
  12. 2025iGaussian: Real-Time Camera Pose Estimation via Feed-Forward 3D Gaussian Splatting Inversioncited by 0