Pith. sign in

REVIEW 3 major objections 5 minor 37 references

GURecon: Learning Detailed 3D Geometric Uncertainties for Neural Surface Reconstruction

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read GURecon learns a per-point 3D geometric uncertainty field purely from multi-view photometric consistency, with no ground-truth geometry, and uses it to pick better next views.

desk verdict A solid, novel method for 3D geometric uncertainty in neural surface reconstruction, with a load-bearing but acknowledged weakness in textureless scenes. read the letter →

arxiv 2412.14939 v3 pith:YKBEDCHX submitted 2024-12-19 cs.CV

Zesong Yang , Ru Zhang , Jiale Shi , Zixiang Ai , Boming Zhao , Hujun Bao , Luwei Yang , Zhaopeng Cui This is my paper · ORCID
classification cs.CV
keywords geometricuncertaintyneuralsurfacereconstructionmulti-viewphotometricconsistencydistillationimplicitsigneddistancefieldsnext-best-viewselectionview-dependentdecoupling3D
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

GURecon sets out to show that a neural surface reconstruction's geometric quality can be scored per point without any ground-truth geometry, by learning a continuous 3D uncertainty field from multi-view photometric consistency. The method locates surface intersection points along rays in an implicit signed-distance field, measures how well small patches around those points agree across views, and trains a position-only uncertainty network to imitate those agreement scores. A decoupled appearance field removes view-dependent lighting and reflections before a final fine-tuning pass, so the uncertainty reflects geometry rather than illumination. If the paper is right, a reconstructed mesh carries its own reliability map, and that map can drive next-best-view selection while remaining independent of the viewing direction.

What carries the argument

The machinery is an online distillation loop from geometry to uncertainty. At each training step, linear interpolation finds the ray segment where the SDF changes sign, giving surface points $T$; homography warping and SSIM over $11\times11$ gray patches produce a consistency score $G^k$ (Eq. 5--6); and the uncertainty MLP $f_u(x)$, which takes only position, is trained with $L_{\mathrm{distill}} = \frac{1}{|R'|}\sum_{r\in R'} |f_u(p'_r) - G_r|$ (Eq. 7). A second decoupled branch writes color as $C = C_{vi} + C_{vd}$, and the rendered view-dependent part is subtracted from input images before the consistency scores are recomputed for a short fine-tuning stage, removing lighting and specular interference from the pseudo-labels.

What would settle it

Reconstruct a scene containing a large uniformly textured wall with GURecon, then displace the extracted surface a few centimeters parallel to the wall; if the uncertainty field assigns low uncertainty to the displaced surface because the photo patches still match, while the distance to the true surface is large, the photometric-consistency proxy is falsified.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that patch-based multi-view photometric consistency at surface points is a usable pseudo-label for geometric reconstruction error, and that a neural field can learn to predict that error in 3D. GURecon finds the zero-crossing of the SDF along each ray, approximates a local tangent plane at the intersection, warps image patches between views with a homography, scores agreement with SSIM on gray patches, keeps the four best-scoring pairs, and distills the resulting score into an uncertainty field $f_u(x)$ that depends only on position. Because the field is view-independent, the same surface point receives the same uncertainty from any direction, which distinguishes it from pixel-wise rendering-based uncertainty. The paper reports that the learned uncertainties align with true distance-to-ground-truth error and improve incremental reconstruction.

Load-bearing premise

The method treats photometric consistency — the same surface patch looking alike across several views — as a sign that the surface is geometrically correct, which fails when a wrong surface sits on blank or repeating texture and still looks consistent.

Editorial extensions

If this is right

  • A scan can be annotated with per-point geometric confidence after training, letting a user see exactly which regions of a reconstructed mesh are trustworthy without comparing to any ground truth.
  • Next-best-view planning gains a view-independent signal: from any candidate camera, the uncertainty field can be rendered by sphere tracing, so the next viewpoint is chosen by summing predicted uncertainty over the visible surface.
  • The distillation schema transfers across surface representations, so new implicit or splatting-based reconstructors can inherit uncertainty estimates without redesigning their architecture.
  • Separating view-dependent appearance from geometry makes uncertainty scores robust to specular highlights and lighting changes across views, a failure mode of photometric-error-based methods.
  • On standardized benchmarks, the reported AUSE3D metric connects 2D image-space error curves to 3D geometric error, giving a direct way to compare uncertainty methods for reconstruction rather than novel-view synthesis.

Reading between the lines

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

  • This suggests that the reliability of the pseudo-labels could be estimated online by the spread among the four selected patch pairs, letting the distillation loss down-weight internally inconsistent matches instead of averaging them.
  • A natural next step is to apply the same consistency-plus-decoupling recipe to pixel-wise rendering uncertainty, which could reduce lighting-induced overconfidence in NeRF-based uncertainty estimators.
  • In active scanning, the continuous field could allow direct gradient-based optimization of the next camera pose, rather than scoring a finite set of candidate views, since uncertainty is defined at every surface point.
  • The method's confidence in a region is bounded by the quality of the current reconstruction, so an incremental loop that re-distills after every added view would need recalibration to avoid compounding early overconfidence.
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

3 major / 5 minor

Summary. The paper proposes GURecon, a framework that learns a continuous 3D geometric uncertainty field for neural implicit surface reconstruction. The method locates zero-crossing surface points of the current SDF, computes patch-based multi-view photometric consistency scores as pseudo-labels, and distills these scores into an uncertainty field via an online distillation loss. A decoupled appearance branch is trained to remove view-dependent factors, and the uncertainty field is fine-tuned on the decoupled images. Experiments on DTU, BlendedMVS, and TNT report improved AUSE and AUSE3D scores relative to adapted NeRF uncertainty baselines, and the uncertainty is used for next-best-view selection in incremental reconstruction, where it improves Chamfer distance and PSNR. The authors claim this is the first geometry-focused, view-independent 3D uncertainty field for neural surfaces, learned without real geometric supervision.

Significance. If the central premise holds, the paper makes a useful contribution: it shifts uncertainty estimation for neural surface reconstruction from rendering-based pixel-wise scores to a 3D field tied to surface geometry, which is more natural for downstream tasks such as next-best-view planning and quality assessment. The plug-and-play extension to 2DGS and the consistent quantitative gains across three datasets are notable strengths, as is the promised code release. The key risk is that the pseudo-label itself is derived from the geometry being learned, so the uncertainty field can be confidently wrong when the surface is consistently wrong but photometrically consistent. The paper's own conclusion concedes degradation on large textureless regions, yet no experiment quantifies this failure mode. These issues are load-bearing for the claim of accurate geometric uncertainty estimation and require additional validation before the result can be fully trusted.

major comments (3)
  1. [3.2-3.3, Eqs. (5)-(7)] The load-bearing assumption is that the patch-based multi-view consistency score G^k is a monotone proxy for the geometric error of the reconstructed surface. Because this score is computed on the currently learned SDF and the uncertainty field is directly trained by L_distill to regress it, a consistently wrong geometry with photometrically consistent projections, such as a shifted low-texture plane or a repetitive-texture region, will receive low uncertainty. Eq. (6) aggregates only the four most consistent pairs, explicitly discarding evidence of inconsistency. The conclusion acknowledges degraded performance on large textureless areas, but the experiments never quantify how often the pseudo-labels are confidently wrong or how the reported AUSE/AUSE3D gains are distributed across textured and textureless regions. I request a direct validation of the pseudo-label: per-point correlation between G^k and the true distance-to-GT error, AUSE/AUSE3D stratified by texture gradient or by consistency-confidence bins, or a synthetic experiment with a known surface shift.
  2. [4.1, Table 2] The baselines are not compared in their native form: ActiveNeRF*, CFNeRF*, and Lee et al.* are structurally modified for SDF backends, while Bayes' Rays and Lee et al. are evaluated as post-hoc frameworks on the model trained by GURecon. As described, this makes the comparison depend on undocumented implementation choices (deferred to the supplementary material) and on the host model's geometry. The claim of superior performance relative to existing methods requires either evaluating all methods on the same SDF geometry with each method's own uncertainty computation, or additionally reporting the uncertainty quality of the original NeRF-based methods on their own reconstructions. Without this, the AUSE improvements cannot be attributed solely to the uncertainty model rather than to the integration strategy or the underlying reconstruction.
  3. [4.1, AUSE metrics] AUSE measures only whether the ranking of predicted uncertainty matches the ranking of error; it does not assess calibration of the predicted uncertainty magnitudes. The qualitative statements in Figs. 3 and 6 that the learned uncertainties align well with GT error are not backed by a quantitative correlation or reliability analysis. Please report a rank correlation (e.g., Spearman) between predicted uncertainty and GT error at the surface-point level, and clarify how AUSE3D is computed, in particular which surface points or rays are used and how the continuous uncertainty field is evaluated on test views. This is needed to support the claim that GURecon models calibrated 3D geometric uncertainty rather than a relative confidence score.
minor comments (5)
  1. [Table 3] The row 'with smaller patch size 70' appears to be a typo; it should read '7' or '7 x 7' to be consistent with the reported patch size of 11 x 11.
  2. [References] The reference 'Stereopsis, R. M. 2010' is incomplete and appears to be a garbled citation for Furukawa and Ponce's multi-view stereo paper; the entry should be corrected and expanded.
  3. [Eq. (8)] The symbol w_r' is used before its definition as the reflection of the view direction around the normal; please define it explicitly at first use and make the prime notation consistent.
  4. [3.5, Implementation Details] The text 'Nf titerations' should read 'N_ft iterations'; the formatting makes the hyperparameter name unclear.
  5. [Figure 5 / Section 4] The caption states that Bayes' Rays and Lee et al. share the same GT curve as ours, but the legend shows only one GT-error curve; please clarify the legend. Also, 'Tanks and Template' should be 'Tanks and Temples', and 'Soilder' in Table 2 should be 'Soldier'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the uncertainty field is trained on a photometric-consistency proxy, but the central evaluation against ground-truth geometric error is external and not reduced to the training signal.

full rationale

The paper's derivation chain is self-contained and not circular. The uncertainty field is supervised via L_distill (Eq. 7) to regress the patch-based multi-view consistency score G computed in Eqs. 5-6. This is a teacher-student distillation setup, not a reduction: the consistency score is a proxy for geometric quality, and the uncertainty field is a learned function of 3D position. The paper's central claim is that this learned uncertainty correlates with true geometric error, which is evaluated against ground-truth depth and meshes via AUSE and AUSE3D on held-out test views. That evaluation is external to the training objective, so the prediction is not equivalent to its input by construction. The acknowledged limitation in the Conclusion—degraded performance in large textureless regions (e.g., white walls)—is a correctness and robustness concern about the proxy's validity, not a circularity. There are no load-bearing self-citations: the cited prior works (Geo-NeuS, NeuS, Bayes' Rays, etc.) are used as standard building blocks or baselines, and the paper does not invoke any author-derived uniqueness theorem or ansatz to force its design. The novelty claims are about modeling a continuous 3D uncertainty field and its downstream use, both of which are substantiated by independent quantitative comparisons. Hence, no circular step meeting the required evidence standard was found.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central claim rests on two hyperparameter groups (patch trade-offs and loss/finetuning choices) and on several modeling assumptions inherited from MVS and reflection decomposition. No new physical entities are introduced; the uncertainty and decoupled fields are learned neural components.

free parameters (5)
  • Patch size for multi-view consistency = 11x11
    Table 3 ablation shows pixel-based and sizes 7 and 15 perform worse; 11x11 is a dataset-specific choice.
  • Number of best patch pairs K = 4
    Eq. (6) selects the four lowest-consistency patch pairs; no principled criterion or ablation for K is provided.
  • Loss weights alpha1..alpha4 = 0.1, 1.0, 0.1, 0.1
    Section 3.5 sets weights manually; no sensitivity analysis is reported.
  • Finetune iterations N_ft = 10k
    Section 3.5: second stage runs 10k additional iterations after 50k training; no ablation on this choice.
  • Sparse training view selection = 6 views for DTU; 25% of images for BlendedMVS/TNT
    Sparse-view setup is chosen to create quality variation; results depend on this split and no sensitivity to view count is reported.
assumptions (6)
  • domain assumption Multi-view photometric consistency is a reliable indicator of geometric reconstruction error.
    Eq. (5)-(6) turn SSIM of warped patches into pseudo-labels G_k that supervise the uncertainty field via L_distill (Eq. 7); the paper does not validate this proxy against GT error, and the conclusion admits failure in large textureless regions.
  • domain assumption Zero-crossing via linear interpolation identifies the true surface point.
    Eq. (2) assumes the SDF changes sign between consecutive samples and is locally linear; thin structures or coarse sampling can violate this.
  • domain assumption The local tangent plane and homography warping model patch projections across views.
    Eq. (3)-(4) rely on the standard planar patch assumption from MVS, which breaks at depth discontinuities and occlusions.
  • domain assumption View-dependent appearance can be decomposed and removed by a learned decoupled field.
    Eq. (8) assumes lighting and reflection are captured by C_vd(n', p', w_r'), following Ref-NeRF-style factorization; errors in this decomposition propagate into the fine-tuned uncertainty.
  • domain assumption Adapting NeRF uncertainty baselines to the SDF representation preserves their intended behavior.
    Section 4.1 modifies ActiveNeRF, CFNeRF, and Lee et al. to surface representations and evaluates Bayes' Rays and Lee et al. post-hoc; the modifications are not fully specified and could bias comparisons.
  • standard math The AUSE and Chamfer Distance metrics are computed consistently with prior work.
    Metrics follow Shen et al. 2022 and Goli et al. 2023; assumes normalization and error definitions are aligned across scenes and methods.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GURecon: Learning Detailed 3D Geometric Uncertainties for Neural Surface Reconstruction." pith.science (2026). https://pith.science/paper/YKBEDCHX

@misc{pith2026241214939,
  author       = {Pith},
  title        = {Pith review of: GURecon: Learning Detailed 3D Geometric Uncertainties for Neural Surface Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YKBEDCHX}},
  note         = {Machine review of arXiv:2412.14939}
}
read the original abstract

Neural surface representation has demonstrated remarkable success in the areas of novel view synthesis and 3D reconstruction. However, assessing the geometric quality of 3D reconstructions in the absence of ground truth mesh remains a significant challenge, due to its rendering-based optimization process and entangled learning of appearance and geometry with photometric losses. In this paper, we present a novel framework, i.e, GURecon, which establishes a geometric uncertainty field for the neural surface based on geometric consistency. Different from existing methods that rely on rendering-based measurement, GURecon models a continuous 3D uncertainty field for the reconstructed surface, and is learned by an online distillation approach without introducing real geometric information for supervision. Moreover, in order to mitigate the interference of illumination on geometric consistency, a decoupled field is learned and exploited to finetune the uncertainty field. Experiments on various datasets demonstrate the superiority of GURecon in modeling 3D geometric uncertainty, as well as its plug-and-play extension to various neural surface representations and improvement on downstream tasks such as incremental reconstruction. The code and supplementary material are available on the project website: https://zju3dv.github.io/GURecon/.

Figures

Figures reproduced from arXiv: 2412.14939 by the authors.

Figure 1
Figure 1. A brief overview. By leveraging multi-view con￾sistency as guidance, GURecon learns detailed 3D geomet￾ric uncertainties for neural surface reconstruction. existing work incorporates uncertainty estimation into NeRF models to identify areas with poor rendering quality. NeRF￾W (Martin-Brualla et al. 2021) and its following works (Pan et al. 2022; Ran et al. 2023) take the radiance field as Gaussian distributions to m… view at source ↗
Figure 2
Figure 2. System Overview. The proposed GURecon models a geometric uncertainty field supervised by the pseudo labels computed based on the multi-view geometry consistency. To deal with the view-dependent factors, additional decoupled fields are also learned and exploited to fine-tune the uncertainty field. With the predicted uncertainty fields, GURecon can boost the downstream tasks such as incremental reconstruction. (Sec. 3… view at source ↗
Figure 4
Figure 4. Uncertainty finetuning with the decoupled fields. Even if the geometry has been well reconstructed, the uncertainty field still erroneously estimates it with high uncertainty caused by light interference across different views. We employ the decoupled fields to remove the view￾dependent factor from the training images. 3.3 Distillation of Geometric Uncertainty Field Considering it is impractical and inefficient to p… view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Sparsification curves of different methods. The dashed and solid lines correspond to the average error of the remaining pixels filtered using GT-error-based and uncertainty-based criteria, the area between them is AUSE. Bayes’ Rays and Lee et al. (Lee et al. 2022) shar…
Figure 7
Figure 7. Figure 7: Qualitative comparison of the uncertainty estimation with 2D depth. Our estimated uncertainties are more closely aligned with the error between the GT and predicted geometry than other baselines (Pan et al. 2022; Shen et al. 2022; Goli et al. 2023; Lee et al. 2022). Fo…
Figure 8
Figure 8. Figure 8: Extension to 2DGS. Our uncertainty distillation can be migrated to various neural surface representations. We show the extension to 2DGS (Huang et al. 2024). trained by our method. From the geometric perspective, although ActiveNeRF* and Bayes’ Rays cannot directly mod…
Figure 9
Figure 9. Figure 9: Incremental results on TNT dataset. Our scheme reconstructs more details while ensuring a smoother surface. Methods Scenes Barn Caterpillar Truck CD(↓) PSNR(↑) CD(↓) PSNR(↑) CD(↓) PSNR(↑) Random 1.033 22.69 0.809 19.71 2.248 20.89 ActiveNeRF* (Pan et al. 2022) 1.002 22…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 26 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    D.; Vogiatzis, G.; Hern \'a ndez, C.; and Cipolla, R

    Campbell, N. D.; Vogiatzis, G.; Hern \'a ndez, C.; and Cipolla, R. 2008. Using multiple hypotheses to improve depth-maps for multi-view stereo. In Eur. Conf. Comput. Vis., 766--779. Springer

    work page 2008

  4. [4]

    Chen, L.; Chen, W.; Wang, R.; and Pollefeys, M. 2023. Leveraging Neural Radiance Fields for Uncertainty-Aware Visual Localization. arXiv preprint arXiv:2310.06984

    work page Pith review arXiv 2023

  5. [5]

    Darmon, F.; Bascle, B.; Devaux, J.-C.; Monasse, P.; and Aubry, M. 2022. Improving neural implicit surfaces geometry with patch warping. In IEEE Conf. Comput. Vis. Pattern Recog., 6260--6269

    work page 2022

  6. [6]

    Fan, Y.; Skorokhodov, I.; Voynov, O.; Ignatyev, S.; Burnaev, E.; Wonka, P.; and Wang, Y. 2023. Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects. arXiv preprint arXiv:2305.17929

    arXiv 2023

  7. [7]

    S.; and Tao, W

    Fu, Q.; Xu, Q.; Ong, Y. S.; and Tao, W. 2022. Geo-neus: Geometry-consistent neural implicit surfaces learning for multi-view reconstruction. Adv. Neural Inform. Process. Syst., 35: 3403--3416

    work page 2022

  8. [8]

    Ge, W.; Hu, T.; Zhao, H.; Liu, S.; and Chen, Y.-C. 2023. Ref-NeuS: Ambiguity-Reduced Neural Implicit Surface Learning for Multi-View Reconstruction with Reflection. arXiv preprint arXiv:2303.10840

    arXiv 2023

Show all 37 references
  1. [9]

    Goli, L.; Reading, C.; Selll \'a n, S.; Jacobson, A.; and Tagliasacchi, A. 2023. Bayes' Rays: Uncertainty Quantification for Neural Radiance Fields. arXiv preprint arXiv:2309.03185

    2023 arXiv

  2. [10]

    Hu, X.; and Mordohai, P. 2012. A Quantitative Evaluation of Confidence Measures for Stereo Vision. IEEE Transactions on Pattern Analysis and Machine Intelligence, 34(11): 2121--2133

    2012

  3. [11]

    Huang, B.; Yu, Z.; Chen, A.; Geiger, A.; and Gao, S. 2024. 2d gaussian splatting for geometrically accurate radiance fields. In ACM SIGGRAPH 2024 Conference Papers, 1--11

    2024

  4. [12]

    Jensen, R.; Dahl, A.; Vogiatzis, G.; Tola, E.; and Aan s, H. 2014. Large scale multi-view stereopsis evaluation. In IEEE Conf. Comput. Vis. Pattern Recog., 406--413

    2014

  5. [13]

    Jiang, W.; Lei, B.; and Daniilidis, K. 2023. FisherRF: Active View Selection and Uncertainty Quantification for Radiance Fields using Fisher Information. arXiv preprint arXiv:2311.17874

    2023 arXiv

  6. [14]

    Knapitsch, A.; Park, J.; Zhou, Q.-Y.; and Koltun, V. 2017. Tanks and temples: Benchmarking large-scale scene reconstruction. ACM Trans. Graph., 36(4): 1--13

    2017

  7. [15]

    Lee, S.; Chen, L.; Wang, J.; Liniger, A.; Kumar, S.; and Yu, F. 2022. Uncertainty guided policy for active robotic 3d reconstruction using neural radiance fields. IEEE Robotics and Automation Letters, 7(4): 12070--12077

    2022

  8. [16]

    Liu, L.; Gu, J.; Zaw Lin, K.; Chua, T.-S.; and Theobalt, C. 2020. Neural sparse voxel fields. Adv. Neural Inform. Process. Syst., 33: 15651--15663

    2020

  9. [17]

    S.; Barron, J

    Martin-Brualla, R.; Radwan, N.; Sajjadi, M. S.; Barron, J. T.; Dosovitskiy, A.; and Duckworth, D. 2021. Nerf in the wild: Neural radiance fields for unconstrained photo collections. In IEEE Conf. Comput. Vis. Pattern Recog., 7210--7219

    2021

  10. [18]

    P.; Tancik, M.; Barron, J

    Mildenhall, B.; Srinivasan, P. P.; Tancik, M.; Barron, J. T.; Ramamoorthi, R.; and Ng, R. 2021. Nerf: Representing scenes as neural radiance fields for view synthesis. Communications of the ACM, 65(1): 99--106

    2021

  11. [19]

    Oechsle, M.; Peng, S.; and Geiger, A. 2021. Unisurf: Unifying neural implicit surfaces and radiance fields for multi-view reconstruction. In Int. Conf. Comput. Vis., 5589--5599

    2021

  12. [20]

    Pan, X.; Lai, Z.; Song, S.; and Huang, G. 2022. Activenerf: Learning where to see with uncertainty estimation. In Eur. Conf. Comput. Vis., 230--246

    2022

  13. [21]

    Ran, Y.; Zeng, J.; He, S.; Chen, J.; Li, L.; Chen, Y.; Lee, G.; and Ye, Q. 2023. NeurAR: Neural Uncertainty for Autonomous 3D Reconstruction With Implicit Neural Representations. IEEE Robotics and Automation Letters, 8(2): 1125--1132

    2023

  14. [22]

    L.; Zheng, E.; Frahm, J.-M.; and Pollefeys, M

    Sch \"o nberger, J. L.; Zheng, E.; Frahm, J.-M.; and Pollefeys, M. 2016. Pixelwise view selection for unstructured multi-view stereo. In Eur. Conf. Comput. Vis., 501--518. Springer

    2016

  15. [23]

    Shen, J.; Agudo, A.; Moreno-Noguer, F.; and Ruiz, A. 2022. Conditional-flow NeRF: Accurate 3D modelling with reliable uncertainty quantification. In Eur. Conf. Comput. Vis., 540--557. Springer

    2022

  16. [24]

    Shen, J.; Ruiz, A.; Agudo, A.; and Moreno-Noguer, F. 2021. Stochastic neural radiance fields: Quantifying uncertainty in implicit 3d representations. In International Conference on 3D Vision (3DV), 972--981. IEEE

    2021

  17. [25]

    Stereopsis, R. M. 2010. Accurate, Dense, and Robust Multiview Stereopsis. IEEE Trans. Pattern Anal. Mach. Intell., 32(8)

    2010

  18. [26]

    S \"u nderhauf, N.; Abou-Chakra, J.; and Miller, D. 2023. Density-aware nerf ensembles: Quantifying predictive uncertainty in neural radiance fields. 9370--9376. IEEE

    2023

  19. [27]

    Tang, J.; Zhou, H.; Chen, X.; Hu, T.; Ding, E.; Wang, J.; and Zeng, G. 2023. Delicate textured mesh recovery from nerf via adaptive surface refinement. arXiv preprint arXiv:2303.02091

    2023 arXiv

  20. [28]

    T.; and Srinivasan, P

    Verbin, D.; Hedman, P.; Mildenhall, B.; Zickler, T.; Barron, J. T.; and Srinivasan, P. P. 2022. Ref-NeRF : Structured View-Dependent Appearance for Neural Radiance Fields. In IEEE Conf. Comput. Vis. Pattern Recog

    2022

  21. [29]

    Wang, J.; Wang, P.; Long, X.; Theobalt, C.; Komura, T.; Liu, L.; and Wang, W. 2022. Neuris: Neural reconstruction of indoor scenes using normal priors. In European Conference on Computer Vision, 139--155. Springer

    2022

  22. [30]

    Wang, P.; Liu, L.; Liu, Y.; Theobalt, C.; Komura, T.; and Wang, W. 2021. Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction. Adv. Neural Inform. Process. Syst

    2021

  23. [31]

    Xiao, Y.; Xu, J.; Yu, Z.; and Gao, S. 2024. Debsdf: Delving into the details and bias of neural indoor scene reconstruction. IEEE Transactions on Pattern Analysis and Machine Intelligence

    2024

  24. [32]

    Yao, Y.; Luo, Z.; Li, S.; Zhang, J.; Ren, Y.; Zhou, L.; Fang, T.; and Quan, L. 2020. Blendedmvs: A large-scale dataset for generalized multi-view stereo networks. In IEEE Conf. Comput. Vis. Pattern Recog., 1790--1799

    2020

  25. [33]

    Yariv, L.; Gu, J.; Kasten, Y.; and Lipman, Y. 2021. Volume rendering of neural implicit surfaces. Adv. Neural Inform. Process. Syst., 34: 4805--4815

    2021

  26. [34]

    Yu, Z.; Peng, S.; Niemeyer, M.; Sattler, T.; and Geiger, A. 2022. Monosdf: Exploring monocular geometric cues for neural implicit surface reconstruction. Adv. Neural Inform. Process. Syst., 35: 25018--25032

    2022

  27. [35]

    Zhan, H.; Zheng, J.; Xu, Y.; Reid, I.; and Rezatofighi, H. 2022. ActiveRMAP: Radiance Field for Active Mapping And Planning. arXiv preprint arXiv:2211.12656

    2022 arXiv

  28. [36]

    Zhang, K.; Riegler, G.; Snavely, N.; and Koltun, V. 2020. Nerf++: Analyzing and improving neural radiance fields. arXiv preprint arXiv:2010.07492

    2020 arXiv

  29. [37]

    Zhao, F.; Jiang, Y.; Yao, K.; Zhang, J.; Wang, L.; Dai, H.; Zhong, Y.; Zhang, Y.; Wu, M.; Xu, L.; et al. 2022. Human performance modeling and rendering via neural animated mesh. ACM Trans. Graph., 41(6): 1--17

    2022

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.