Pith. sign in

REVIEW 4 major objections 6 minor 73 references

Robust SG-NeRF: Robust Scene Graph Aided Neural Surface Reconstruction

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A second, blind color network can flag outlier camera poses, and using that signal to resample rays and re-localize the outliers markedly improves neural surface reconstruction.

desk verdict A credible incremental advance over SG-NeRF with a nice confidence-signal idea; the pose-accuracy results are inflated by a self-referential evaluation, but the reconstruction gains look real. read the letter →

arxiv 2411.13620 v1 pith:MQIDI7V3 submitted 2024-11-20 cs.CV

Yi Gu , Dongjun Ye , Zhaorui Wang , Jiaxu Wang , Jiahang Cao , Renjing Xu This is my paper · ORCID
classification cs.CV
keywords neuralsurfacereconstructioncameraposeoutliersscenegraphshape-radianceambiguityview-direction-freecolornetworkMonteCarlore-localizationsigneddistancefunctionpose-NeRFjointoptimization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Neural surface reconstruction typically assumes camera poses are correct, but poses from COLMAP or ARKit can be badly wrong, and joint pose-scene optimization handles small errors yet fails on large ones such as mirrored views. This paper argues that a rendering-quality gap that standard PSNR misses can be exposed by training a second color network that ignores viewing direction: when the two color networks disagree strongly, the pose is an outlier. The method uses that signal to sample more rays from inlier poses, applies re-projection and intersection-over-union losses to tighten inlier geometry, and runs Monte Carlo re-localization on outliers, with the scene graph updated as training proceeds. On the SG-NeRF benchmark the approach lowers mean Chamfer distance from 0.46 to 0.32 and improves pose errors by over two orders of magnitude, while also improving DTU results.

What carries the argument

The load-bearing object is the detached view-direction-free color network $C_n$, a shallow color MLP that shares sampling points and geometry features with the standard color network $C_o$ but never receives the viewing direction and is cut off from the main optimization graph. Its rendered PSNR, $PSNR_n$, replaces the usual $PSNR_o$ in the confidence update, and the rule $|PSNR_o - PSNR_n| > \tau_1 = 9$ labels a pose as outlier. Confidence scores from the scene graph are updated as $CS(v_i) = CS(v_i) + \lambda_c PSNR_n(v_i)$, so rays are drawn preferentially from inlier poses. Around this indicator the method builds two correction paths: inlier pairs get re-projection and intersection-over-union losses, while outlier poses are re-initialized by particle poses $R_{axis}(i \cdot 2\pi/N_p) \cdot R_o$ about the coarse main axis and optimized with $PSNR_n$ as the selection score.

What would settle it

Take a held-out set of poses with known errors and plot $|PSNR_o - PSNR_n|$ against the true pose error after training. If any pose with a large true error stays below the $\tau_1 = 9$ line, or any accurate pose crosses it, the indicator misfires. A concrete failure mode would be a textureless or repetitive scene where both color networks render confidently from wrong poses, collapsing the gap.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that shape-radiance ambiguity breaks PSNR as a pose-confidence signal: a NeRF with a badly wrong pose can still render images with high PSNR, so inliers and outliers are confused. The authors show that a color network which omits viewing direction, detached from the main pose-NeRF optimization graph, renders the outlier images much worse, and the absolute gap $|PSNR_o - PSNR_n|$ between the standard and detached networks separates the two classes. With this indicator, the framework identifies inliers and outliers, concentrates ray sampling on inlier poses, strengthens inlier matching with re-projection and IoU losses, and re-localizes outlier poses through Monte Carlo sampling around the scene's main axis. The result is consistently better surface geometry and pose accuracy than the scene-graph baseline it builds on, for example mean Chamfer distance 0.32 versus 0.46 on SG-NeRF and absolute pose error 0.053 versus 1.44.

Load-bearing premise

The whole pipeline rests on the assumption that a fixed gap of 9 between the two PSNRs cleanly separates good poses from bad ones in every scene; if that gap mislabels poses, ray sampling and re-localization are pointed at the wrong images.

Editorial extensions

If this is right

  • Up to roughly a third of input poses can be badly wrong and the reconstruction still recovers high-fidelity geometry in one training stage.
  • Mirrored or otherwise severely wrong poses are corrected rather than discarded, so all captured images contribute to the model.
  • The same confidence signal drives both ray sampling and outlier re-localization, making the two mechanisms a single feedback loop.
  • Pose accuracy improves by more than two orders of magnitude over the scene-graph baseline on the SG-NeRF benchmark, which would make downstream AR and robotics tasks far more reliable.
  • Dynamic scene graph updating prunes wrong keypoint matches during training, strengthening constraints as geometry improves.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The fixed threshold $\tau_1 = 9$ invites a calibration study: plotting the two-PSNR gap against known pose error per scene would reveal whether one threshold survives across datasets or should be tuned per scene.
  • Because the indicator only needs two color heads, it could be grafted onto other NeRF-based surface backbones, not just NeuS, as a plug-in outlier detector.
  • The method's acknowledged dependence on a substantial number of inlier poses suggests it would degrade in sparse-view settings, so prior-based pose priors would be a direct extension.
  • The PSNR gap may also serve as an unsupervised signal for detecting mirror-symmetry ambiguity in structure-from-motion itself.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This paper proposes RSG-NeRF, a neural surface reconstruction method built on NeuS that aims to handle both small pose noise and large outlier poses. The method uses a scene graph to obtain initial per-image confidence scores, adds a detached color network that omits viewing direction, and classifies poses as inliers or outliers based on the PSNR gap between the standard color network and the view-direction-free network with a fixed threshold tau1 = 9. It then samples rays preferentially from inlier poses, applies re-projection and IoU losses to inlier-inlier pairs, employs Monte Carlo re-localization for outlier poses, and periodically updates the scene graph by filtering keypoint matches with a re-projection threshold. Experiments on the SG-NeRF and DTU datasets report improved Chamfer distance, F-score, APE, and RPE compared with SG-NeRF and other baselines.

Significance. If the reported gains hold, the paper makes a meaningful contribution to robust neural surface reconstruction under severely noisy camera poses, particularly for mirror-symmetry ambiguity. The paper's strengths include a clear problem formulation, a novel detached color network for confidence estimation, a reasonably detailed ablation study, and a public project page with visual results. However, the central claim of superior pose accuracy currently rests on an evaluation that filters poses using the method's own classifier, and the outlier detector itself is not validated against known labels or threshold sensitivity. These issues must be resolved before the significance claim becomes fully credible.

major comments (4)
  1. [§4.3, pose accuracy evaluation] The APE/RPE results in Table 1 are computed after filtering out outliers using the proposed indicator (|PSNR_o − PSNR_n| > τ1), as stated in Sec. 4.3: 'we utilize our indicator to filter out outliers.' Because the same indicator determines which poses are optimized as inliers and which are re-localized, the reported pose error measures the method's self-consistency rather than an independent accuracy. This is especially problematic when the same filter is applied to SG-NeRF poses, since SG-NeRF does not use this confidence model. Please recompute APE/RPE on all poses without filtering, or use ground-truth inlier/outlier labels (available for the DTU noise-injection protocol) to select poses for the SIM(3) alignment; otherwise the claim of improvement 'by more than two orders of magnitude' is not supported.
  2. [§3.1, Eq. (2) and threshold τ1] The inlier/outlier classification is load-bearing: it gates confidence-based ray sampling, the re-projection and IoU losses, and Monte Carlo re-localization. Yet the classifier is justified only by a qualitative example in Fig. 2 and by favorable end results; no ROC analysis, confusion matrix, or threshold sweep is provided. The DTU experiments inject known outlier poses (Sec. 4.1) and therefore permit a direct label-based evaluation of the detector, but this is not reported. Please add a sensitivity analysis for τ1 (and ideally for the loss weights α, β, λ) and report per-scene classification accuracy or an ROC curve on the labeled DTU outliers; this is necessary to establish that the confidence signal, rather than dataset-specific behavior, drives the improvements.
  3. [§4.1 and §4.3, DTU experiments] The paper states that Monte Carlo re-localization 'has not been triggered' on the DTU dataset, so the DTU experiments do not validate the outlier-handling mechanism that is a central contribution. The DTU results therefore only demonstrate an improved SG-NeRF with additional geometric losses. Since the authors claim a robust solution for outlier poses, the DTU evaluation should either inject sufficiently large noise to trigger re-localization, or the paper should explicitly state that the re-localization component remains unvalidated on DTU. Please also clarify in the contributions which components are supported by which experiments.
  4. [Abstract and Sec. 4.3, Tables 1–2] The abstract and Sec. 4.3 state that the method 'consistently' outperforms SG-NeRF, but Tables 1 and 2 show two counterexamples: the Bell scene in Table 1 (Chamfer distance 1.22 vs. 0.98 for SG-NeRF) and Scan 55 in Table 2 (0.44 vs. 0.38 for SG-NeRF). Please qualify the claim to reflect scene-dependent performance or provide an explanation for these regressions.
minor comments (6)
  1. [§3.1] The phrase 'donated as P SN Ro' should be 'denoted as PSNR_o'.
  2. [Fig. 6 caption] The caption contains a typo: 'The fist row' should be 'The first row'.
  3. [References [13] and [36]] The reference list entries for [13] and [36] end with stray page/citation markers ', 1' that appear to be formatting artifacts.
  4. [§3.2, Eqs. (6)–(8)] The term 'Intersection-of-Union' is more commonly rendered as 'Intersection-over-Union'; consider standardizing the terminology.
  5. [§4.3, Table 2 discussion] The sentence 'In DTU [6] dataset' should reference the DTU dataset [25], not [6].
  6. [§3.2, Eq. (9)] The Huber loss Lδ is not fully specified; please state the Huber parameter in the implementation details.

Circularity Check

1 steps flagged · score 3.0 of 10

Pose-accuracy evaluation is filtered by the method's own unvalidated confidence indicator, making the headline APE/RPE comparison partly self-referential; reconstruction metrics remain independent.

  1. other [Section 4.3, pose accuracy evaluation (Table 1)]
    "Due to the original SG-NeRF dataset does not provide inlier-outlier information, we utilize our indicator to filter out outliers. We align inliers to ground truth poses to get a global SIM(3) transformation, which is then applied to all poses. The results of absolute pose error (APE) and relative pose error (RPE) w.r.t. full transformation (including both rotation and translation parts) are reported in Table 1. Our pose accuracy surpasses that of SG-NeRF by more than two orders of magnitude on both RPE and APE."

    The pose-accuracy claim is computed after excluding poses flagged by the paper's own confidence indicator (|PSNR_o - PSNR_n| > tau1, Sec. 3.1), which is exactly the component whose effectiveness the paper aims to demonstrate. The SIM(3) alignment used for APE/RPE is therefore anchored on poses selected by the method under test rather than on an independent inlier/outlier ground truth. Any error in the indicator changes which poses define the alignment, so the reported 'surpasses ... by more than two orders of magnitude' is conditional on the correctness of the very detector being proposed.

full rationale

The core reconstruction results (Chamfer distance and F-score on SG-NeRF and DTU) are evaluated against ground-truth geometry and are not derived from the proposed confidence scheme, so the main reconstruction claim has independent content. The confidence indicator itself is presented as an empirical heuristic with a fixed threshold tau1=9 and is not a fitted parameter that directly forces the output metrics. However, the pose-accuracy evaluation explicitly reuses the method's own indicator to select inlier poses for the SIM(3) alignment, as quoted above. Since the dataset provides no independent inlier/outlier labels, this creates a feedback loop in the reported APE/RPE numbers: the evaluation anchor depends on the same classifier whose performance is being claimed. This is a genuine but partial circularity, and it is compounded by the paper's admission that Monte Carlo re-localization was never triggered on DTU, so the DTU experiments do not independently validate the outlier-handling machinery. Overall, the reconstruction-quality comparison is self-contained and non-circular, while the pose-accuracy comparison is partially self-referential; hence the score is 3 rather than higher.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The method has four fitted hyperparameters and relies on standard NeuS/SfM machinery. The key ad hoc assumption is the reliability of the detached view-direction-free color network's PSNR for pose confidence.

free parameters (4)
  • inlier-outlier PSNR threshold τ1 = 9
    Set empirically to distinguish inlier and outlier poses based on the PSNR gap between the two color networks.
  • angular threshold τ = 70 (SG-NeRF), 45 (DTU)
    Filters scene graph edges by relative rotation to remove incorrect matches.
  • loss weights α, β, λ = 0.2, 0.001, 0.1
    Weights for IoU loss, reprojection loss, and SDF regularization, chosen without sensitivity analysis.
  • particle number Np = 24
    Number of Monte Carlo re-localization particles, set for efficiency.
assumptions (3)
  • domain assumption NeuS SDF-based volume rendering accurately models geometry
    The framework builds on NeuS, assuming its rendering pipeline is a reliable backbone for surface reconstruction.
  • domain assumption Scene graph from COLMAP provides correct matching pairs and initial poses
    The entire method relies on the quality of the initial scene graph; the paper only sparsifies it with thresholds, not re-estimating matches from scratch.
  • ad hoc to paper PSNR from a view-direction-free color network is a reliable confidence signal
    The paper empirically observes this but provides no theoretical justification; this is the central premise of the method.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robust SG-NeRF: Robust Scene Graph Aided Neural Surface Reconstruction." pith.science (2026). https://pith.science/paper/MQIDI7V3

@misc{pith2026241113620,
  author       = {Pith},
  title        = {Pith review of: Robust SG-NeRF: Robust Scene Graph Aided Neural Surface Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MQIDI7V3}},
  note         = {Machine review of arXiv:2411.13620}
}
read the original abstract

Neural surface reconstruction relies heavily on accurate camera poses as input. Despite utilizing advanced pose estimators like COLMAP or ARKit, camera poses can still be noisy. Existing pose-NeRF joint optimization methods handle poses with small noise (inliers) effectively but struggle with large noise (outliers), such as mirrored poses. In this work, we focus on mitigating the impact of outlier poses. Our method integrates an inlier-outlier confidence estimation scheme, leveraging scene graph information gathered during the data preparation phase. Unlike previous works directly using rendering metrics as the reference, we employ a detached color network that omits the viewing direction as input to minimize the impact caused by shape-radiance ambiguities. This enhanced confidence updating strategy effectively differentiates between inlier and outlier poses, allowing us to sample more rays from inlier poses to construct more reliable radiance fields. Additionally, we introduce a re-projection loss based on the current Signed Distance Function (SDF) and pose estimations, strengthening the constraints between matching image pairs. For outlier poses, we adopt a Monte Carlo re-localization method to find better solutions. We also devise a scene graph updating strategy to provide more accurate information throughout the training process. We validate our approach on the SG-NeRF and DTU datasets. Experimental results on various datasets demonstrate that our methods can consistently improve the reconstruction qualities and pose accuracies.

Figures

Figures reproduced from arXiv: 2411.13620 by the authors.

Figure 1
Figure 1. Reconstruction results on the SG-NeRF [6] dataset. Both SG-NeRF [6] and our method take the same initial poses as input, including significant noises. The camera poses are also presented with optimized outlier poses, inlier poses and ground truth poses. More results are illustrated in the supplementary material. through a signed distance function (SDF), which is learned by a multilayer perceptron (MLP) network train… view at source ↗
Figure 2
Figure 2. The illustration of the pose ambiguity. The first row is [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. An overview of the proposed pipeline. Given the initial scene graph, we apply a confidence updating strategy based on an [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: The illustration of scene graph updating. We filter out [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparisons on the SG-NeRF [ [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Visualization of pose accuracy. The fist row presents [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Qualitative comparison on the DTU [25] dataset. L2G-NeRF [5] is trained in a two-stage manner and others are trained in one stage with the same iterations. All methods take the same initial poses as input [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: and 9 illustrate more results of the scene graph updating. We present 4 cases in Bear and Baby scenes, in￾cluding all rejected, all accepted, more inlier matching and more outlier matching in each [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: Scene graph updating on Baby. 1 [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: Reconstruction results on the SG-NeRF [6] dataset. Both SG-NeRF [6] and our method take the same initial poses as input, including significant noises. The camera poses are also presented with optimized outlier poses, inlier poses and ground truth poses. Reference SG-N…
Figure 11
Figure 11. Figure 11: Reconstruction results on the SG-NeRF [6] dataset (Baby, Bear). 2 [PITH_FULL_IMAGE:figures/full_fig_p013_11.png]
Figure 12
Figure 12. Figure 12: Reconstruction results on the SG-NeRF [6] dataset (Bell, Clock). Reference SG-NeRF Ours Reference SG-NeRF Ours [PITH_FULL_IMAGE:figures/full_fig_p014_12.png]
Figure 13
Figure 13. Figure 13: Reconstruction results on the SG-NeRF [6] dataset (Deaf, Farmer). Reference SG-NeRF Ours Reference SG-NeRF Ours [PITH_FULL_IMAGE:figures/full_fig_p014_13.png]
Figure 14
Figure 14. Figure 14: Reconstruction results on the SG-NeRF [6] dataset (Pavilion, Sculpture). 3 [PITH_FULL_IMAGE:figures/full_fig_p014_14.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

73 extracted references · 47 canonical work pages

  1. [1]

    Patchmatch: A randomized correspon- dence algorithm for structural image editing

    Connelly Barnes, Eli Shechtman, Adam Finkelstein, and Dan B Goldman. Patchmatch: A randomized correspon- dence algorithm for structural image editing. ACM Trans. Graph., 28(3):24, 2009. 2

    2009

  2. [2]

    Porf: Pose residual field for accurate neural sur- face reconstruction

    Jia-Wang Bian, Wenjing Bian, Victor Adrian Prisacariu, and Philip Torr. Porf: Pose residual field for accurate neural sur- face reconstruction. In ICLR, 2024. 1, 3, 4, 6, 7, 8

    work page 2024

  3. [3]

    Nope-nerf: Optimising neu- ral radiance field with no pose prior

    Wenjing Bian, Zirui Wang, Kejie Li, Jia-Wang Bian, and Victor Adrian Prisacariu. Nope-nerf: Optimising neu- ral radiance field with no pose prior. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4160–4169, 2023. 1, 3

    work page 2023

  4. [4]

    Visual camera re- localization from rgb and rgb-d images using dsac

    Eric Brachmann and Carsten Rother. Visual camera re- localization from rgb and rgb-d images using dsac. IEEE transactions on pattern analysis and machine intelligence , 44(9):5847–5865, 2021. 3

    work page 2021

  5. [5]

    Local-to-global registration for bundle-adjusting neural radiance fields

    Yue Chen, Xingyu Chen, Xuan Wang, Qi Zhang, Yu Guo, Ying Shan, and Fei Wang. Local-to-global registration for bundle-adjusting neural radiance fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8264–8273, 2023. 1, 3, 6, 7, 8

    work page 2023

  6. [6]

    Sg-nerf: Neural surface recon- struction with scene graph optimization

    Yiyang Chen, Siyan Dong, Xulong Wang, Lulu Cai, Youyi Zheng, and Yanchao Yang. Sg-nerf: Neural surface recon- struction with scene graph optimization. In European Con- ference on Computer Vision (ECCV), 2024. 1, 2, 3, 4, 5, 6, 7, 8

    work page 2024

  7. [7]

    Improving Robustness for Joint Optimization of Camera Poses and Decomposed Low-Rank Tensorial Radiance Fields

    Bo-Yu Cheng, Wei-Chen Chiu, and Yu-Lun Liu. Improv- ing robustness for joint optimization of camera poses and de- composed low-rank tensorial radiance fields. arXiv preprint arXiv:2402.13252, 2024. 3, 6, 7, 8

    work page Pith review arXiv 2024

  8. [8]

    Lu-nerf: Scene and pose estimation by synchronizing local unposed nerfs

    Zezhou Cheng, Carlos Esteves, Varun Jampani, Abhishek Kar, Subhransu Maji, and Ameesh Makadia. Lu-nerf: Scene and pose estimation by synchronizing local unposed nerfs. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 18312–18321, 2023. 1, 5

    work page 2023

Show all 73 references
  1. [9]

    Gaussian activated neural radiance fields for high fidelity reconstruction and pose estimation

    Shin-Fang Chng, Sameera Ramasinghe, Jamie Sherrah, and Simon Lucey. Gaussian activated neural radiance fields for high fidelity reconstruction and pose estimation. In Eu- ropean Conference on Computer Vision , pages 264–280. Springer, 2022. 1, 3

    2022

  2. [10]

    Gaussian activated neural radiance fields for high fidelity reconstruction and pose estimation

    Shin-Fang Chng, Sameera Ramasinghe, Jamie Sherrah, and Simon Lucey. Gaussian activated neural radiance fields for high fidelity reconstruction and pose estimation. In Eu- ropean Conference on Computer Vision , pages 264–280. Springer, 2022. 3, 6, 7, 8

    2022

  3. [11]

    Invertible neural warp for nerf

    Shin-Fang Chng, Ravi Garg, Hemanth Saratchandran, and Simon Lucey. Invertible neural warp for nerf. In European Conference on Computer Vision (ECCV), 2024. 3

    2024

  4. [12]

    Global structure-from-motion by similarity averaging

    Zhaopeng Cui and Ping Tan. Global structure-from-motion by similarity averaging. In Proceedings of the IEEE Inter- national Conference on Computer Vision , pages 864–872,

  5. [13]

    Monte carlo localization for mobile robots

    Frank Dellaert, Dieter Fox, Wolfram Burgard, and Sebas- tian Thrun. Monte carlo localization for mobile robots. In Proceedings 1999 IEEE international conference on robotics and automation (Cat. No. 99CH36288C), pages 1322–1328. IEEE, 1999. 5, 1

    1999

  6. [14]

    Superpoint: Self-supervised interest point detection and description

    Daniel DeTone, Tomasz Malisiewicz, and Andrew Rabi- novich. Superpoint: Self-supervised interest point detection and description. In Proceedings of the IEEE conference on computer vision and pattern recognition workshops , pages 224–236, 2018. 3, 6

    2018

  7. [15]

    Robust neural routing through space partitions for camera relocal- ization in dynamic indoor environments

    Siyan Dong, Qingnan Fan, He Wang, Ji Shi, Li Yi, Thomas Funkhouser, Baoquan Chen, and Leonidas J Guibas. Robust neural routing through space partitions for camera relocal- ization in dynamic indoor environments. In Proceedings of the IEEE/CVF Conference on Computer Vision and ...

    2021

  8. [16]

    D2-net: A trainable CNN for joint description and detection of local features

    Mihai Dusmanu, Ignacio Rocco, Tom ´as Pajdla, Marc Polle- feys, Josef Sivic, Akihiko Torii, and Torsten Sattler. D2-net: A trainable CNN for joint description and detection of local features. In IEEE Conference on Computer Vision and Pat- tern Recognition, CVPR 2019, Long Beac...

    2019

  9. [17]

    Re- ducing shape-radiance ambiguity in radiance fields with a closed-form color estimation method

    Qihang Fang, Yafei Song, Keqiang Li, and Liefeng Bo. Re- ducing shape-radiance ambiguity in radiance fields with a closed-form color estimation method. Advances in Neural Information Processing Systems, 36, 2024. 2, 4

    2024

  10. [18]

    Geo-neus: Geometry-consistent neural implicit surfaces learning for multi-view reconstruction

    Qiancheng Fu, Qingshan Xu, Yew Soon Ong, and Wenbing Tao. Geo-neus: Geometry-consistent neural implicit surfaces learning for multi-view reconstruction. Advances in Neural Information Processing Systems, 35:3403–3416, 2022. 2

    2022

  11. [19]

    Accurate, dense, and robust multiview stereopsis

    Yasutaka Furukawa and Jean Ponce. Accurate, dense, and robust multiview stereopsis. IEEE transactions on pattern analysis and machine intelligence , 32(8):1362–1376, 2009. 2

    2009

  12. [20]

    Gipuma: Massively parallel multi-view stereo reconstruc- tion

    Silvano Galliani, Katrin Lasinger, and Konrad Schindler. Gipuma: Massively parallel multi-view stereo reconstruc- tion. Publikationen der Deutschen Gesellschaft f ¨ur Pho- togrammetrie, Fernerkundung und Geoinformation e. V , 25 (361-369):2, 2016. 2

    2016

  13. [21]

    Implicit geometric regularization for learning shapes

    Amos Gropp, Lior Yariv, Niv Haim, Matan Atzmon, and Yaron Lipman. Implicit geometric regularization for learning shapes. arXiv preprint arXiv:2002.10099, 2020. 4

    2002 arXiv

  14. [22]

    evo: Python package for the evalua- tion of odometry and slam

    Michael Grupp. evo: Python package for the evalua- tion of odometry and slam. https://github.com/ MichaelGrupp/evo, 2017. 7

    2017

  15. [23]

    Multiple view ge- ometry in computer vision

    Richard Hartley and Andrew Zisserman. Multiple view ge- ometry in computer vision . Cambridge university press,

  16. [24]

    Robust camera pose refinement for multi-resolution hash encoding

    Hwan Heo, Taekyung Kim, Jiyoung Lee, Jaewon Lee, Soohyun Kim, Hyunwoo J Kim, and Jin-Hwa Kim. Robust camera pose refinement for multi-resolution hash encoding. arXiv preprint arXiv:2302.01571, 2023. 3 9

    2023 arXiv

  17. [25]

    Large scale multi-view stereopsis eval- uation

    Rasmus Jensen, Anders Dahl, George V ogiatzis, Engin Tola, and Henrik Aanæs. Large scale multi-view stereopsis eval- uation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 406–413, 2014. 2, 6, 8

    2014

  18. [26]

    Self-calibrating neural radiance fields

    Yoonwoo Jeong, Seokjun Ahn, Christopher Choy, Anima Anandkumar, Minsu Cho, and Jaesik Park. Self-calibrating neural radiance fields. In Proceedings of the IEEE/CVF In- ternational Conference on Computer Vision , pages 5846– 5854, 2021. 1, 3, 6, 7, 8

    2021

  19. [27]

    Ray tracing volume densities

    James T Kajiya and Brian P V on Herzen. Ray tracing volume densities. ACM SIGGRAPH computer graphics, 18(3):165– 174, 1984. 4

    1984

  20. [28]

    Screened poisson sur- face reconstruction

    Michael Kazhdan and Hugues Hoppe. Screened poisson sur- face reconstruction. ACM Transactions on Graphics (ToG), 32(3):1–13, 2013. 2

    2013

  21. [29]

    Neuralangelo: High-fidelity neural surface reconstruction

    Zhaoshuo Li, Thomas M ¨uller, Alex Evans, Russell H Tay- lor, Mathias Unberath, Ming-Yu Liu, and Chen-Hsuan Lin. Neuralangelo: High-fidelity neural surface reconstruction. In IEEE Conference on Computer Vision and Pattern Recogni- tion (CVPR), 2023. 1, 2, 6, 7, 8

    2023

  22. [30]

    Barf: Bundle-adjusting neural radiance fields

    Chen-Hsuan Lin, Wei-Chiu Ma, Antonio Torralba, and Si- mon Lucey. Barf: Bundle-adjusting neural radiance fields. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 5741–5751, 2021. 1, 3, 6, 7, 8

    2021

  23. [31]

    Pixel-Perfect Structure-from-Motion with Featuremetric Refinement

    Philipp Lindenberger, Paul-Edouard Sarlin, Viktor Larsson, and Marc Pollefeys. Pixel-Perfect Structure-from-Motion with Featuremetric Refinement. In ICCV, 2021. 3

    2021

  24. [32]

    Lightglue: Local feature matching at light speed

    Philipp Lindenberger, Paul-Edouard Sarlin, and Marc Polle- feys. Lightglue: Local feature matching at light speed. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 17627–17638, 2023. 3

    2023

  25. [33]

    Nerf- loc: Visual localization with conditional neural radiance field

    Jianlin Liu, Qiang Nie, Yong Liu, and Chengjie Wang. Nerf- loc: Visual localization with conditional neural radiance field. In 2023 IEEE International Conference on Robotics and Automation (ICRA), pages 9385–9392. IEEE, 2023. 3

    2023

  26. [34]

    3d line mapping revisited

    Shaohui Liu, Yifan Yu, R ´emi Pautrat, Marc Pollefeys, and Viktor Larsson. 3d line mapping revisited. In Computer Vision and Pattern Recognition (CVPR), 2023. 3

    2023

  27. [35]

    Distinctive image features from scale- invariant keypoints

    David G Lowe. Distinctive image features from scale- invariant keypoints. International journal of computer vi- sion, 60:91–110, 2004. 3

    2004

  28. [36]

    Loc-nerf: Monte carlo local- ization using neural radiance fields

    Dominic Maggio, Marcus Abate, Jingnan Shi, Courtney Mario, and Luca Carlone. Loc-nerf: Monte carlo local- ization using neural radiance fields. In 2023 IEEE Inter- national Conference on Robotics and Automation (ICRA) , pages 4018–4025. IEEE, 2023. 3, 5, 1

    2023

  29. [37]

    Real-time visibility-based fusion of depth maps

    Paul Merrell, Amir Akbarzadeh, Liang Wang, Philippos Mordohai, Jan-Michael Frahm, Ruigang Yang, David Nist´er, and Marc Pollefeys. Real-time visibility-based fusion of depth maps. In 2007 IEEE 11th International Conference on Computer Vision, pages 1–8. Ieee, 2007. 2

    2007

  30. [38]

    Srinivasan, Matthew Tancik, Jonathan T

    Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view syn- thesis. In ECCV, 2020. 2, 3, 4

    2020

  31. [39]

    Nerf: Representing scenes as neural radiance fields for view syn- thesis

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view syn- thesis. Communications of the ACM , 65(1):99–106, 2021. 1

    2021

  32. [40]

    Objects as volumes: A stochastic geometry view of opaque solids

    Bailey Miller, Hanyu Chen, Alice Lai, and Ioannis Gkioulekas. Objects as volumes: A stochastic geometry view of opaque solids. InProceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 87–97, 2024. 1

    2024

  33. [41]

    Crossfire: Camera relocalization on self-supervised fea- tures from an implicit representation

    Arthur Moreau, Nathan Piasco, Moussab Bennehar, Dzmitry Tsishkou, Bogdan Stanciulescu, and Arnaud de La Fortelle. Crossfire: Camera relocalization on self-supervised fea- tures from an implicit representation. arXiv preprint arXiv:2303.04869, 2023. 3

    2023 arXiv

  34. [42]

    Iterative pose estimation using coplanar feature points.Com- puter Vision and Image Understanding , 63(3):495–511,

    Denis Oberkampf, Daniel F DeMenthon, and Larry S Davis. Iterative pose estimation using coplanar feature points.Com- puter Vision and Image Understanding , 63(3):495–511,

  35. [43]

    Multibody structure-from-motion in practice

    Kemal E Ozden, Konrad Schindler, and Luc Van Gool. Multibody structure-from-motion in practice. IEEE Trans- actions on Pattern Analysis and Machine Intelligence, 32(6): 1134–1141, 2010. 5

    2010

  36. [44]

    Global Structure-from-Motion Revisited

    Linfei Pan, Daniel Barath, Marc Pollefeys, and Jo- hannes Lutz Sch ¨onberger. Global Structure-from-Motion Revisited. In European Conference on Computer Vision (ECCV), 2024. 3

    2024

  37. [45]

    Camp: Camera preconditioning for neural radiance fields

    Keunhong Park, Philipp Henzler, Ben Mildenhall, Jonathan T Barron, and Ricardo Martin-Brualla. Camp: Camera preconditioning for neural radiance fields. ACM Transactions on Graphics (TOG), 42(6):1–11, 2023. 3

    2023

  38. [46]

    R2d2: Reliable and repeatable detec- tor and descriptor

    Jerome Revaud, Cesar De Souza, Martin Humenberger, and Philippe Weinzaepfel. R2d2: Reliable and repeatable detec- tor and descriptor. Advances in neural information process- ing systems, 32, 2019. 3

    2019

  39. [47]

    From coarse to fine: Robust hierarchical localization at large scale

    Paul-Edouard Sarlin, Cesar Cadena, Roland Siegwart, and Marcin Dymczyk. From coarse to fine: Robust hierarchical localization at large scale. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 12716–12725, 2019. 3, 6

    2019

  40. [48]

    SuperGlue: Learning feature matching with graph neural networks

    Paul-Edouard Sarlin, Daniel DeTone, Tomasz Malisiewicz, and Andrew Rabinovich. SuperGlue: Learning feature matching with graph neural networks. In CVPR, 2020. 3, 6

    2020

  41. [49]

    Efficient & effective prioritized matching for large-scale image-based localization

    Torsten Sattler, Bastian Leibe, and Leif Kobbelt. Efficient & effective prioritized matching for large-scale image-based localization. IEEE transactions on pattern analysis and ma- chine intelligence, 39(9), 2016. 3

    2016

  42. [50]

    Structure-from-motion revisited

    Johannes Lutz Sch ¨onberger and Jan-Michael Frahm. Structure-from-motion revisited. In Conference on Com- puter Vision and Pattern Recognition (CVPR) , 2016. 1, 2, 3

    2016

  43. [51]

    Pixelwise view selection for un- structured multi-view stereo

    Johannes Lutz Sch ¨onberger, Enliang Zheng, Marc Pollefeys, and Jan-Michael Frahm. Pixelwise view selection for un- structured multi-view stereo. In European Conference on Computer Vision (ECCV), 2016. 1, 2

    2016

  44. [52]

    Robust pose estimation from a planar target

    Gerald Schweighofer and Axel Pinz. Robust pose estimation from a planar target. IEEE transactions on pattern analysis and machine intelligence, 28(12):2024–2030, 2006. 5 10

    2024

  45. [53]

    Model- ing the world from internet photo collections

    Noah Snavely, Steven M Seitz, and Richard Szeliski. Model- ing the world from internet photo collections. International journal of computer vision, 80:189–210, 2008. 3

    2008

  46. [54]

    Loftr: Detector-free local feature matching with transformers

    Jiaming Sun, Zehong Shen, Yuang Wang, Hujun Bao, and Xiaowei Zhou. Loftr: Detector-free local feature matching with transformers. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 8922–8931, 2021. 3

    2021

  47. [55]

    Theia multiview geometry library: Tutorial & reference

    Chris Sweeney. Theia multiview geometry library: Tutorial & reference. http://theia-sfm.org. 3

  48. [56]

    Tangent space backpropa- gation for 3d transformation groups

    Zachary Teed and Jia Deng. Tangent space backpropa- gation for 3d transformation groups. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10338–10347, 2021. 6

    2021

  49. [57]

    Sparf: Neural radiance fields from sparse and noisy poses

    Prune Truong, Marie-Julie Rakotosaona, Fabian Manhardt, and Federico Tombari. Sparf: Neural radiance fields from sparse and noisy poses. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 4190–4200, 2023. 1, 3

    2023

  50. [58]

    Disk: Learning local features with policy gradient

    Michał Tyszkiewicz, Pascal Fua, and Eduard Trulls. Disk: Learning local features with policy gradient. Advances in Neural Information Processing Systems , 33:14254–14265,

  51. [59]

    Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction

    Peng Wang, Lingjie Liu, Yuan Liu, Christian Theobalt, Taku Komura, and Wenping Wang. Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction. NeurIPS, 2021. 1, 2, 3, 4, 6, 7, 8

    2021

  52. [60]

    Neus2: Fast learning of neural implicit surfaces for multi-view recon- struction

    Yiming Wang, Qin Han, Marc Habermann, Kostas Dani- ilidis, Christian Theobalt, and Lingjie Liu. Neus2: Fast learning of neural implicit surfaces for multi-view recon- struction. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3295–3306, 2023. 2

    2023

  53. [62]

    Nerf–: Neural radiance fields without known camera parameters

    Zirui Wang, Shangzhe Wu, Weidi Xie, Min Chen, and Victor Adrian Prisacariu. Nerf–: Neural radiance fields without known camera parameters. arXiv preprint arXiv:2102.07064, 2021. 1, 3, 4, 5

    2021 arXiv

  54. [63]

    Visualsfm: A visual structure from motion system

    Changchang Wu. Visualsfm: A visual structure from motion system. http://www. cs. washington. edu/homes/ccwu/vsfm ,

  55. [64]

    V oxurf: V oxel-based efficient and accurate neural surface reconstruction

    Tong Wu, Jiaqi Wang, Xingang Pan, Xudong Xu, Christian Theobalt, Ziwei Liu, and Dahua Lin. V oxurf: V oxel-based efficient and accurate neural surface reconstruction. arXiv preprint arXiv:2208.12697, 2022. 1, 2

    2022 arXiv

  56. [65]

    Sinerf: Sinusoidal neural radiance fields for joint pose estimation and scene reconstruction

    Yitong Xia, Hao Tang, Radu Timofte, and Luc Van Gool. Sinerf: Sinusoidal neural radiance fields for joint pose estimation and scene reconstruction. arXiv preprint arXiv:2210.04553, 2022. 3

    2022 arXiv

  57. [66]

    Mvsnet: Depth inference for unstructured multi-view stereo

    Yao Yao, Zixin Luo, Shiwei Li, Tian Fang, and Long Quan. Mvsnet: Depth inference for unstructured multi-view stereo. In Proceedings of the European conference on computer vi- sion (ECCV), pages 767–783, 2018. 2

    2018

  58. [67]

    Recurrent mvsnet for high-resolution multi-view stereo depth inference

    Yao Yao, Zixin Luo, Shiwei Li, Tianwei Shen, Tian Fang, and Long Quan. Recurrent mvsnet for high-resolution multi-view stereo depth inference. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5525–5534, 2019. 2

    2019

  59. [68]

    Multiview neu- ral surface reconstruction by disentangling geometry and ap- pearance

    Lior Yariv, Yoni Kasten, Dror Moran, Meirav Galun, Matan Atzmon, Basri Ronen, and Yaron Lipman. Multiview neu- ral surface reconstruction by disentangling geometry and ap- pearance. Advances in Neural Information Processing Sys- tems, 33, 2020. 3

    2020

  60. [69]

    V ol- ume rendering of neural implicit surfaces

    Lior Yariv, Jiatao Gu, Yoni Kasten, and Yaron Lipman. V ol- ume rendering of neural implicit surfaces. Advances in Neu- ral Information Processing Systems, 34:4805–4815, 2021. 2

    2021

  61. [70]

    inerf: Inverting neural radiance fields for pose estimation

    Lin Yen-Chen, Pete Florence, Jonathan T Barron, Alberto Rodriguez, Phillip Isola, and Tsung-Yi Lin. inerf: Inverting neural radiance fields for pose estimation. In2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1323–1330. IEEE, 2021. 3

    2021

  62. [71]

    Monosdf: Exploring monocu- lar geometric cues for neural implicit surface reconstruc- tion

    Zehao Yu, Songyou Peng, Michael Niemeyer, Torsten Sat- tler, and Andreas Geiger. Monosdf: Exploring monocu- lar geometric cues for neural implicit surface reconstruc- tion. Advances in neural information processing systems , 35:25018–25032, 2022. 2

    2022

  63. [72]

    A glob- ally optimal algorithm for robust tv-l 1 range image integra- tion

    Christopher Zach, Thomas Pock, and Horst Bischof. A glob- ally optimal algorithm for robust tv-l 1 range image integra- tion. In 2007 IEEE 11th International Conference on Com- puter Vision, pages 1–8. IEEE, 2007. 2

    2007

  64. [73]

    Nerf++: Analyzing and improving neural radiance fields

    Kai Zhang, Gernot Riegler, Noah Snavely, and Vladlen Koltun. Nerf++: Analyzing and improving neural radiance fields. arXiv:2010.07492, 2020. 2, 4

    2010 arXiv

  65. [74]

    Vdn-nerf: Resolving shape-radiance ambiguity via view-dependence normalization

    Bingfan Zhu, Yanchao Yang, Xulong Wang, Youyi Zheng, and Leonidas Guibas. Vdn-nerf: Resolving shape-radiance ambiguity via view-dependence normalization. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 35–45, 2023. 2, 4 11 Robust S...

    2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.