pith. machine review for the scientific record. sign in

arxiv: 2604.11992 · v1 · submitted 2026-04-13 · 💻 cs.RO · cs.CV

ReefMapGS: Enabling Large-Scale Underwater Reconstruction by Closing the Loop Between Multimodal SLAM and Gaussian Splatting

Daniel Yang , Jungseok Hong , John J. Leonard , Yogesh Girdhar This is my paper

Pith reviewed 2026-05-10 15:37 UTC · model grok-4.3

classification 💻 cs.RO cs.CV
keywords underwater reconstruction3D Gaussian Splattingmultimodal SLAMAUV mappingreef sitespose graph optimizationincremental reconstructionCOLMAP-free
0
0 comments X

The pith

ReefMapGS closes the loop between multimodal SLAM and 3D Gaussian Splatting to produce large-scale underwater reef models without structure-from-motion.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents ReefMapGS as an incremental reconstruction method that starts from a high-certainty region of the scene and grows outward by alternating between adding new image observations and optimizing the underlying 3D Gaussian Splatting representation. Refined camera poses from this optimization step are sent back into a multimodal pose-graph optimizer that fuses acoustic, inertial, pressure, and visual measurements. The process repeats across the full trajectory, yielding both a dense scene model and an improved global vehicle path. This matters for field robotics because traditional structure-from-motion pipelines are too slow and brittle for underwater conditions with variable lighting and scale. The approach demonstrates results on two real reef sites with trajectories reaching 700 meters.

Core claim

ReefMapGS builds an initial model from a high certainty region and progressively expands to incorporate the whole scene by interleaving local tracking of new image observations with optimization of the underlying 3DGS scene; these refined poses are integrated back into the pose-graph to globally optimize the whole trajectory, resulting in COLMAP-free 3D reconstruction of underwater reef sites with complex geometry as well as more accurate global pose estimation of the AUV over survey trajectories spanning up to 700 m.

What carries the argument

The closed feedback loop in which 3D Gaussian Splatting optimization refines camera poses that are then used to update the multimodal pose-graph, allowing incremental expansion from an initial high-certainty seed region.

If this is right

  • Underwater sites with complex geometry can be reconstructed at field scale without offline structure-from-motion processing.
  • AUV pose estimates improve in global consistency when visual scene optimization is interleaved with multimodal pose-graph updates.
  • Incremental model growth supports continuous operation over trajectories hundreds of meters long.
  • Multimodal uncertainty estimates from the SLAM layer become usable for guiding the Gaussian Splatting optimization.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same interleaving pattern could be tested in other sensor-rich but visually degraded settings such as murky water or low-light caves.
  • Quantitative comparison against ground-truth trajectories on additional reef datasets would clarify how much the feedback loop reduces drift.
  • Extending the method to streaming reconstruction might allow an AUV to maintain an up-to-date map during a single dive.
  • The framework suggests that dense visual representations can serve as an external consistency check for any pose-graph SLAM system.

Load-bearing premise

That poses refined by 3D Gaussian Splatting can be fed back into the pose-graph optimizer to increase global consistency without causing divergence or new drift under variable underwater lighting and geometry.

What would settle it

A full reef survey trajectory processed end-to-end where the final global trajectory error or reconstruction completeness is worse after the pose feedback step than when using only the original multimodal SLAM poses.

Figures

Figures reproduced from arXiv: 2604.11992 by Daniel Yang, John J. Leonard, Jungseok Hong, Yogesh Girdhar.

Figure 1
Figure 1. Figure 1: ReefMapGS integrates multimodal pose-graph SLAM with 3D Gaussian Splatting to enable rapid, dense reconstruction of underwater environments like coral reefs without needing computationally expensive structure-from-motion pipelines. On the top left, we show an image captured by a snorkeler on the water surface passively observing a robot performing a visual benthic survey of a coral reef. Combining the AUV-… view at source ↗
Figure 2
Figure 2. Figure 2: System overview ReefMapGStakes as input RGB images and inertial, acoustic, and pressure sensor data from an AUV. Sensor data is fused into odometry while the known, central landmark is detected from RGB frames. Together, this information is integrated into a factor graph to estimate the whole AUV trajectory, shown in the center. We incrementally build a 3DGS model of the scene, starting from the region of … view at source ↗
Figure 3
Figure 3. Figure 3: Incremental, radially-expanding 3DGS optimization framework. Local camera pose refinement occurs as we radially and incrementally expand our 3DGS-based map representation, incorporating new cameras and their visual data into the scene. Given a current 3DGS map, target image, and initial camera pose estimate, we can refine the camera pose with gradient descent (as shown from the light to dark frustums). Sam… view at source ↗
Figure 4
Figure 4. Figure 4: Qualitative trajectory evaluation Top-down view of robot tra￾jectories estimated by ReefMapGSand other baselines methods are plotted against the reference trajectory estimated by Metashape, dashed green, at both reef sites, Tektite (above) and Yawzi (below). Metashape+sfm+batch. When incorporating our incremental optimization with local pose refinement as new frontiers are explored (+inc instead of +batch)… view at source ↗
read the original abstract

3D Gaussian Splatting is a powerful visual representation, providing high-quality and efficient 3D scene reconstruction, but it is crucially dependent on accurate camera poses typically obtained from computationally intensive processes like structure-from-motion that are unsuitable for field robot applications. However, in these domains, multimodal sensor data from acoustic, inertial, pressure, and visual sensors are available and suitable for pose-graph optimization-based SLAM methods that can estimate the vehicle's trajectory and thus our needed camera poses while providing uncertainty. We propose a 3DGS-based incremental reconstruction framework, ReefMapGS, that builds an initial model from a high certainty region and progressively expands to incorporate the whole scene. We reconstruct the scene incrementally by interleaving local tracking of new image observations with optimization of the underlying 3DGS scene. These refined poses are integrated back into the pose-graph to globally optimize the whole trajectory. We show COLMAP-free 3D reconstruction of two underwater reef sites with complex geometry as well as more accurate global pose estimation of our AUV over survey trajectories spanning up to 700 m.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces ReefMapGS, a framework for large-scale underwater 3D reconstruction that combines multimodal SLAM with 3D Gaussian Splatting. It initializes from a high-certainty region and incrementally expands the reconstruction by interleaving local tracking of new images with 3DGS optimization. Refined poses from this process are fed back into the multimodal pose-graph optimization to improve global consistency. The authors claim this enables COLMAP-free reconstruction of two complex reef sites and more accurate AUV pose estimation over trajectories up to 700 m.

Significance. If the closed-loop integration between 3DGS and multimodal SLAM proves robust, this work could significantly advance field-deployable underwater mapping by leveraging readily available sensor modalities to achieve high-quality reconstructions without reliance on computationally heavy offline SfM methods like COLMAP. This is particularly valuable for AUV operations in challenging marine environments where accurate large-scale models are needed for ecological monitoring or navigation.

major comments (2)
  1. [Abstract] Abstract: The central claim of 'more accurate global pose estimation' over 700 m trajectories and successful COLMAP-free reconstruction of two reef sites is presented without any quantitative metrics, baselines, error bars, ablation studies, or uncertainty quantification. This absence is load-bearing because the soundness of the closed-loop improvement cannot be assessed from the given description alone.
  2. [Abstract] Abstract: The incremental expansion process is described as feeding 3DGS-refined poses back into the pose-graph, but no explicit mechanism (e.g., uncertainty-aware weighting, selective insertion, or robust loss) is mentioned to guard against divergence from underwater photometric errors such as scattering, attenuation, or non-Lambertian surfaces. This directly bears on the weakest assumption that the loop reliably improves rather than corrupts global consistency.
minor comments (2)
  1. The abstract would be clearer if it briefly specified the exact multimodal sensors (acoustic, inertial, pressure, visual) and how their uncertainties are modeled in the initial pose-graph.
  2. A high-level diagram of the interleaving between local 3DGS tracking/optimization and global pose-graph update would improve readability of the incremental pipeline.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We agree that the abstract requires quantitative metrics to support the central claims and an explicit mention of the integration safeguards. We will revise the abstract accordingly while ensuring the manuscript body already provides the supporting details and experiments.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim of 'more accurate global pose estimation' over 700 m trajectories and successful COLMAP-free reconstruction of two reef sites is presented without any quantitative metrics, baselines, error bars, ablation studies, or uncertainty quantification. This absence is load-bearing because the soundness of the closed-loop improvement cannot be assessed from the given description alone.

    Authors: We agree that the abstract would benefit from including key quantitative results to substantiate the claims. The full manuscript reports these evaluations in the experiments, including pose error reductions versus baselines over the 700 m trajectories and reconstruction metrics for the two reef sites. We will revise the abstract to summarize these quantitative findings, such as the reported accuracy improvements and COLMAP-free outcomes. revision: yes

  2. Referee: [Abstract] Abstract: The incremental expansion process is described as feeding 3DGS-refined poses back into the pose-graph, but no explicit mechanism (e.g., uncertainty-aware weighting, selective insertion, or robust loss) is mentioned to guard against divergence from underwater photometric errors such as scattering, attenuation, or non-Lambertian surfaces. This directly bears on the weakest assumption that the loop reliably improves rather than corrupts global consistency.

    Authors: The manuscript details the feedback mechanism in the methods section, where 3DGS-refined poses are incorporated into the pose-graph optimization using uncertainty estimates from the multimodal SLAM to provide weighting that mitigates the impact of underwater photometric errors. We acknowledge the abstract does not explicitly reference this safeguard. We will revise the abstract to briefly note the uncertainty-aware integration and strengthen the methods discussion on robustness to scattering, attenuation, and non-Lambertian effects. revision: partial

Circularity Check

0 steps flagged

No significant circularity; closed-loop method relies on empirical interaction of independent components

full rationale

The paper describes an incremental framework that starts with multimodal SLAM poses from acoustic/inertial/pressure/visual data, builds an initial 3DGS model in a high-certainty region, interleaves local tracking and scene optimization, and feeds refined poses back into global pose-graph optimization. No equations, derivations, or fitted parameters are shown that reduce the claimed pose accuracy or reconstruction quality to quantities defined by the method itself. The central claim is an empirical assertion about the closed-loop behavior on real 700 m reef trajectories, not a tautological re-expression of inputs. No self-citations, uniqueness theorems, or ansatzes are invoked in the provided text to justify load-bearing steps.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities; assessment is limited to the high-level description.

pith-pipeline@v0.9.0 · 5502 in / 1090 out tokens · 31405 ms · 2026-05-10T15:37:53.821720+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

46 extracted references · 46 canonical work pages · 1 internal anchor

  1. [1]

    Curee: A curious un- derwater robot for ecosystem exploration,

    Y . Girdhar, N. McGuire, L. Cai, S. Jamieson, S. McCammon, B. Claus, J. E. S. Soucie, J. E. Todd, and T. A. Mooney, “Curee: A curious un- derwater robot for ecosystem exploration,” in2023 IEEE International Conference on Robotics and Automation, 2023, pp. 11 411–11 417

    work page 2023

  2. [2]

    Underwater exploration and mapping,

    B. Joshi, M. Xanthidis, M. Roznere, N. J. Burgdorfer, P. Mordohai, A. Q. Li, and I. Rekleitis, “Underwater exploration and mapping,” in 2022 IEEE/OES Autonomous Underwater V ehicles Symposium (AUV). IEEE, 2022, pp. 1–7

    work page 2022

  3. [3]

    High-resolution underwater robotic vision-based mapping and three-dimensional reconstruction for ar- chaeology,

    M. Johnson-Roberson, M. Bryson, A. Friedman, O. Pizarro, G. Troni, P. Ozog, and J. C. Henderson, “High-resolution underwater robotic vision-based mapping and three-dimensional reconstruction for ar- chaeology,”Journal of Field Robotics, vol. 34, no. 4, pp. 625–643, 2017

    work page 2017

  4. [4]

    Advancements in sensor fusion for underwater slam: A review on enhanced navigation and environmental perception,

    F. F. R. Merveille, B. Jia, Z. Xu, and B. Fred, “Advancements in sensor fusion for underwater slam: A review on enhanced navigation and environmental perception,”Sensors (Basel, Switzerland), vol. 24, no. 23, p. 7490, 2024

    work page 2024

  5. [5]

    Simultaneous localisation and mapping on the great barrier reef,

    S. Williams and I. Mahon, “Simultaneous localisation and mapping on the great barrier reef,” inIEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA ’04. 2004, vol. 2, 2004, pp. 1771–1776 V ol.2

    work page 2004

  6. [6]

    3d gaussian splatting for real-time radiance field rendering

    B. Kerbl, G. Kopanas, T. Leimk ¨uhler, G. Drettakiset al., “3d gaussian splatting for real-time radiance field rendering.”ACM Trans. Graph., vol. 42, no. 4, pp. 139–1, 2023

    work page 2023

  7. [7]

    Splatam: Splat, track and map 3d gaussians for dense rgb-d slam,

    N. Keetha, J. Karhade, K. M. Jatavallabhula, G. Yang, S. Scherer, D. Ramanan, and J. Luiten, “Splatam: Splat, track and map 3d gaussians for dense rgb-d slam,” in2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 21 357– 21 366

    work page 2024

  8. [8]

    Wildgs-slam: Monocular gaussian splatting slam in dynamic envi- ronments,

    J. Zheng, Z. Zhu, V . Bieri, M. Pollefeys, S. Peng, and I. Armeni, “Wildgs-slam: Monocular gaussian splatting slam in dynamic envi- ronments,” in2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025, pp. 11 461–11 471

    work page 2025

  9. [9]

    Gaussian Splatting SLAM,

    H. Matsuki, R. Murai, P. H. J. Kelly, and A. J. Davison, “Gaussian Splatting SLAM,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

    work page 2024

  10. [10]

    Dust3r: Geometric 3d vision made easy,

    S. Wang, V . Leroy, Y . Cabon, B. Chidlovskii, and J. Revaud, “Dust3r: Geometric 3d vision made easy,” in2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 20 697–20 709

    work page 2024

  11. [11]

    Must3r: Multi-view network for stereo 3d reconstruc- tion,

    Y . Cabon, L. Stoffl, L. Antsfeld, G. Csurka, B. Chidlovskii, J. Revaud, and V . Leroy, “Must3r: Multi-view network for stereo 3d reconstruc- tion,” inProceedings of the Computer Vision and Pattern Recognition Conference, 2025, pp. 1050–1060

    work page 2025

  12. [12]

    Mast3r-sfm: a fully-integrated solution for unconstrained structure-from-motion,

    B. P. Duisterhof, L. Zust, P. Weinzaepfel, V . Leroy, Y . Cabon, and J. Revaud, “Mast3r-sfm: a fully-integrated solution for unconstrained structure-from-motion,” in2025 International Conference on 3D Vi- sion (3DV). IEEE, 2025, pp. 1–10

    work page 2025

  13. [13]

    Continuous 3d perception model with persistent state,

    Q. Wang, Y . Zhang, A. Holynski, A. A. Efros, and A. Kanazawa, “Continuous 3d perception model with persistent state,” in2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025, pp. 10 510–10 522

    work page 2025

  14. [14]

    3D Reconstruction with Spatial Memory

    H. Wang and L. Agapito, “3d reconstruction with spatial memory,” arXiv preprint arXiv:2408.16061, 2024

    work page internal anchor Pith review arXiv 2024

  15. [15]

    Vggt: Visual geometry grounded transformer,

    J. Wang, M. Chen, N. Karaev, A. Vedaldi, C. Rupprecht, and D. Novotny, “Vggt: Visual geometry grounded transformer,” in2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025, pp. 5294–5306

    work page 2025

  16. [16]

    A survey of underwater vehicle navigation: Recent advances and new challenges,

    J. C. Kinsey, R. M. Eustice, and L. L. Whitcomb, “A survey of underwater vehicle navigation: Recent advances and new challenges,” inIF AC conference of manoeuvering and control of marine craft, vol. 88. Lisbon, 2006, pp. 1–12

    work page 2006

  17. [17]

    Autonomous underwater vehicle naviga- tion,

    J. J. Leonard and A. Bahr, “Autonomous underwater vehicle naviga- tion,”Springer handbook of ocean engineering, pp. 341–358, 2016

    work page 2016

  18. [18]

    Experimental comparison of open source visual-inertial-based state estimation algorithms in the underwater domain,

    B. Joshi, S. Rahman, M. Kalaitzakis, B. Cain, J. Johnson, M. Xan- thidis, N. Karapetyan, A. Hernandez, A. Q. Li, N. Vitzilaioset al., “Experimental comparison of open source visual-inertial-based state estimation algorithms in the underwater domain,” in2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2019, pp. 7227–7233

    work page 2019

  19. [19]

    Svin2: An underwater slam system using sonar, visual, inertial, and depth sensor,

    S. Rahman, A. Q. Li, and I. Rekleitis, “Svin2: An underwater slam system using sonar, visual, inertial, and depth sensor,” in2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2019, pp. 1861–1868

    work page 2019

  20. [20]

    Real-time dense 3d mapping of underwater environments,

    W. Wang, B. Joshi, N. Burgdorfer, K. Batsosc, A. Q. Lid, P. Mordo- haia, and I. Rekleitisb, “Real-time dense 3d mapping of underwater environments,” in2023 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2023, pp. 5184–5191

    work page 2023

  21. [21]

    Aqua-slam: Tightly-coupled un- derwater acoustic-visual-inertial slam with sensor calibration,

    S. Xu, K. Zhang, and S. Wang, “Aqua-slam: Tightly-coupled un- derwater acoustic-visual-inertial slam with sensor calibration,”IEEE Transactions on Robotics, 2025

    work page 2025

  22. [22]

    Turtlmap: Real-time localization and dense mapping of low-texture underwater environments with a low-cost unmanned underwater vehi- cle,

    J. Song, O. Bagoren, R. Andigani, A. Sethuraman, and K. A. Skinner, “Turtlmap: Real-time localization and dense mapping of low-texture underwater environments with a low-cost unmanned underwater vehi- cle,” in2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024, pp. 1191–1198

    work page 2024

  23. [23]

    Structure-from-motion revisited,

    J. L. Sch ¨onberger and J.-M. Frahm, “Structure-from-motion revisited,” inConference on Computer Vision and Pattern Recognition, 2016

    work page 2016

  24. [24]

    Agisoft Metashape: Professional Edition

    “Agisoft Metashape: Professional Edition.” [Online]. Available: https://www.agisoft.com/features/professional-edition/

    work page

  25. [25]

    Scalable semantic 3d mapping of coral reefs with deep learning,

    J. Sauder, G. Banc-Prandi, A. Meibom, and D. Tuia, “Scalable semantic 3d mapping of coral reefs with deep learning,”Methods in Ecology and Evolution, vol. 15, no. 5, pp. 916–934, 2024

    work page 2024

  26. [26]

    Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Video,

    J. Bian, Z. Li, N. Wang, H. Zhan, C. Shen, M.-M. Cheng, and I. Reid, “Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Video,” inAdvances in Neural Information Processing Systems, vol. 32, 2019

    work page 2019

  27. [27]

    Nerf: Representing scenes as neural radiance fields for view synthesis,

    B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoor- thi, and R. Ng, “Nerf: Representing scenes as neural radiance fields for view synthesis,” inECCV, 2020

    work page 2020

  28. [28]

    Keeping a good attitude: A quaternion-based orientation filter for imus and margs,

    R. G. Valenti, I. Dryanovski, and J. Xiao, “Keeping a good attitude: A quaternion-based orientation filter for imus and margs,”Sensors, vol. 15, no. 8, pp. 19 302–19 330, 2015

    work page 2015

  29. [29]

    A generalized extended kalman filter imple- mentation for the robot operating system,

    T. Moore and D. Stouch, “A generalized extended kalman filter imple- mentation for the robot operating system,” inIntelligent Autonomous Systems 13: Proceedings of the 13th International Conference IAS-13. Springer, 2016, pp. 335–348

    work page 2016

  30. [30]

    gsplat: An open-source library for gaussian splatting,

    V . Ye, R. Li, J. Kerr, M. Turkulainen, B. Yi, Z. Pan, O. Seiskari, J. Ye, J. Hu, M. Tanciket al., “gsplat: An open-source library for gaussian splatting,”Journal of Machine Learning Research, vol. 26, no. 34, pp. 1–17, 2025

    work page 2025

  31. [31]

    Unsupervised monocu- lar depth estimation with left-right consistency,

    C. Godard, O. Mac Aodha, and G. J. Brostow, “Unsupervised monocu- lar depth estimation with left-right consistency,” inProceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 270–279

    work page 2017

  32. [32]

    Dinov2: Learning robust visual features without supervision,

    M. Oquab, T. Darcet, T. Moutakanni, H. V . V o, M. Szafraniec, V . Khalidov, P. Fernandez, D. Haziza, F. Massa, A. El-Nouby, R. Howes, P.-Y . Huang, H. Xu, V . Sharma, S.-W. Li, W. Galuba, M. Rabbat, M. Assran, N. Ballas, G. Synnaeve, I. Misra, H. Jegou, J. Mairal, P. Labatut, A. Joulin, and P. Bojanowski, “Dinov2: Learning robust visual features without ...

    work page 2023

  33. [33]

    WildGaussians: 3D gaussian splatting in the wild,

    J. Kulhanek, S. Peng, Z. Kukelova, M. Pollefeys, and T. Sattler, “WildGaussians: 3D gaussian splatting in the wild,”Advances in Neural Information Processing Systems, vol. 38, 2024

    work page 2024

  34. [34]

    Nerf on-the-go: Exploiting uncertainty for distractor-free nerfs in the wild,

    W. Ren, Z. Zhu, B. Sun, J. Chen, M. Pollefeys, and S. Peng, “Nerf on-the-go: Exploiting uncertainty for distractor-free nerfs in the wild,” in2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 8931–8940

    work page 2024

  35. [35]

    Improving 2D Feature Representations by 3D-Aware Fine-Tuning,

    Y . Yue, A. Das, F. Engelmann, S. Tang, and J. E. Lenssen, “Improving 2D Feature Representations by 3D-Aware Fine-Tuning,” inEuropean Conference on Computer Vision (ECCV), 2024

    work page 2024

  36. [36]

    3D Gaussian Splatting as Markov Chain Monte Carlo,

    S. Kheradmand, D. Rebain, G. Sharma, W. Sun, Y .-C. Tseng, H. Isack, A. Kar, A. Tagliasacchi, and K. M. Yi, “3D Gaussian Splatting as Markov Chain Monte Carlo,”Advances in Neural Information Processing Systems, vol. 38, 2024

    work page 2024

  37. [37]

    Depth anything v2,

    L. Yang, B. Kang, Z. Huang, Z. Zhao, X. Xu, J. Feng, and H. Zhao, “Depth anything v2,”Advances in Neural Information Processing Systems, vol. 37, 2024

    work page 2024

  38. [38]

    Dn-splatter: Depth and normal priors for gaussian splatting and meshing,

    M. Turkulainen, X. Ren, I. Melekhov, O. Seiskari, E. Rahtu, and J. Kannala, “Dn-splatter: Depth and normal priors for gaussian splatting and meshing,” in2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2025, pp. 2421–2431

    work page 2025

  39. [39]

    Replayed reef sounds induce settlement of favia fragum coral larvae in aquaria and field environments,

    N. Aoki, B. Weiss, Y . J ´ez´equel, A. Apprill, and T. A. Mooney, “Replayed reef sounds induce settlement of favia fragum coral larvae in aquaria and field environments,”JASA Express Letters, vol. 4, no. 10, 2024

    work page 2024

  40. [40]

    AprilTag: A robust and flexible visual fiducial system,

    E. Olson, “AprilTag: A robust and flexible visual fiducial system,” in Proceedings of the IEEE International Conference on Robotics and Automation (ICRA). IEEE, May 2011, pp. 3400–3407

    work page 2011

  41. [41]

    evo: Python package for the evaluation of odometry and slam

    M. Grupp, “evo: Python package for the evaluation of odometry and slam.” https://github.com/MichaelGrupp/evo, 2017

    work page 2017

  42. [42]

    Orb-slam3: An accurate open-source library for visual, visual–inertial, and multimap slam,

    C. Campos, R. Elvira, J. J. G. Rodr ´ıguez, J. M. Montiel, and J. D. Tard ´os, “Orb-slam3: An accurate open-source library for visual, visual–inertial, and multimap slam,”IEEE transactions on robotics, vol. 37, no. 6, pp. 1874–1890, 2021

    work page 2021

  43. [43]

    DROID-SLAM: Deep Visual SLAM for Monoc- ular, Stereo, and RGB-D Cameras,

    Z. Teed and J. Deng, “DROID-SLAM: Deep Visual SLAM for Monoc- ular, Stereo, and RGB-D Cameras,”Advances in neural information processing systems, 2021

    work page 2021

  44. [44]

    MASt3R-SLAM: Real- time dense SLAM with 3D reconstruction priors,

    R. Murai, E. Dexheimer, and A. J. Davison, “MASt3R-SLAM: Real- time dense SLAM with 3D reconstruction priors,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2025

    work page 2025

  45. [45]

    Vggt-slam: Dense rgb slam optimized on the sl (4) manifold,

    D. Maggio, H. Lim, and L. Carlone, “Vggt-slam: Dense rgb slam optimized on the sl (4) manifold,”Advances in Neural Information Processing Systems, vol. 39, 2025

    work page 2025

  46. [46]

    borglab/gtsam,

    F. Dellaert and G. Contributors, “borglab/gtsam,” May 2022. [Online]. Available: https://github.com/borglab/gtsam)

    work page 2022