Pith. sign in

REVIEW 4 major objections 4 minor 62 references

FreBIS: Frequency-Based Stratification for Neural Implicit Surface Representations

T0 review · 4 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read FreBIS claims that splitting scene frequencies across three specialist encoders, weighted by mutual distinctness, improves reconstructed 3D surfaces and rendering for an off-the-shelf method like VolSDF on BlendedMVS.

desk verdict FreBIS is a clean frequency-stratified encoder variant of VolSDF with small rendering gains and an unverified surface-quality claim; the missing quantitative geometry evaluation is the main gap. read the letter →

arxiv 2504.20222 v1 pith:54KZOIKA submitted 2025-04-28 cs.CV

classification cs.CV
keywords neuralimplicitsurfacessigneddistancefunctionsfrequency-basedstratificationpositionalencodingmulti-view3Dreconstructionvolumerenderingredundancy-awareweightingBlendedMVS
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

The paper proposes FreBIS, a way to rebuild the input encoder of a neural implicit surface model so that low-, middle-, and high-frequency surface details are learned by three separate encoders rather than one shared network. Its central claim is that swapping this frequency-stratified encoder into a standard method, VolSDF, improves both the fidelity of rendered views and the visual quality of the reconstructed 3D mesh on the BlendedMVS dataset, at the same parameter count. A second claim is that a redundancy-aware weighting module, which up-weights each encoder's feature to the extent it differs from the other two, helps the encoders encode complementary information and is responsible for a measurable share of the gain. If the claims hold, the result suggests that one cause of detail loss in implicit surface reconstruction is encoder-capacity conflict across frequencies, and that splitting the frequency spectrum is a fix that can be dropped into existing decoders.

What carries the argument

The load-bearing object is the frequency-stratified encoder plus the redundancy-aware weighting rule. The input coordinate is mapped by a standard positional encoding $\gamma(x)$ into a vector of sines and cosines at levels $2^0$ through $2^{N-1}$; FreBIS partitions these levels into bands and routes each band to its own MLP encoder, so no single network must explain both macroscopic shape and microscopic texture. The weighting module computes $F = [f_L, f_M, f_H]$, L2-normalizes each column, forms the Gram-style similarity matrix $S = \bar{F}^T \bar{F}$, removes the diagonal with $S' = S - I$, and sets the dissimilarity vector $d = (2I - S')\cdot \mathbf{1}$, then applies $w = \mathrm{Softmax}(d/\tau)$ with default $\tau=0.5$. This gives higher weight to the feature column least predicted by the others and lower weight to redundant columns, which is what pushes the encoders to specialize; the weighted features $F\cdot\mathrm{diag}(w)$ feed a VolSDF-style decoder that outputs an SDF value and an RGB appearance feature.

What would settle it

Run FreBIS and its baseline on a multi-view benchmark that provides ground-truth meshes, and compare Chamfer distance and F-score between the reconstructed surfaces; if FreBIS's rendering metrics improve while its geometric accuracy matches or trails the baseline, the paper's surface-reconstruction claim is refuted, and if both improve, it is confirmed.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the single-encoder design of neural implicit surface methods, not their decoder or volume-rendering machinery, is the bottleneck for scenes containing both smooth structure and fine detail. FreBIS stratifies the positional encoding of a 3D point into three non-overlapping frequency bands, two octaves per band when the total is N=6, and dedicates one 256-dimensional encoder to each band. A redundancy-aware weighting module then forms a normalized feature matrix, computes pairwise cosine similarities between the three feature columns, subtracts self-similarity, and derives a softmax weight for each encoder that grows with its dissimilarity to the other two; the weighted combination is decoded into a signed distance value and an appearance feature. The paper reports that this encoder replacement outperforms VolSDF and a parameter-matched scaled-up VolSDF on BlendedMVS on PSNR, SSIM, and LPIPS in nearly all tested scenes, and that the reconstructed meshes preserve details such as a doll's dress bands and a robot's gun that the baselines blur or lose. It also reports that the frequency bands behave as intended: decoded per-band meshes show low frequencies carrying global shape, middle frequencies rough structure, and high frequencies fine carving, while the weighting visualizations place high-frequency features on detailed regions.

Load-bearing premise

The load-bearing assumption is that better rendering metrics and visual mesh inspection are enough to show the reconstructed 3D surface is better, because BlendedMVS has no ground-truth mesh to measure geometry directly.

Editorial extensions

If this is right

  • Replacing a single implicit-surface encoder with frequency-band specialists improves rendered image quality and mesh detail on BlendedMVS, and the paper's scaled-up VolSDF baseline shows the gain is not simply extra parameters.
  • Assigning the six frequency levels evenly across the three encoders is the most stable configuration across scenes, whereas uneven splits help some scenes but hurt others.
  • FreBIS remains usable at higher total frequency levels (N=9, 12) where a scaled-up single encoder degrades or fails to produce a mesh, because the bands shield low-frequency shape information from high-frequency noise.
  • Ablating the redundancy-aware weighting and averaging the features instead lowers PSNR, SSIM, and LPIPS on the Dog scene, so the dissimilarity weighting contributes beyond mere stratification.
  • Since the decoder is off-the-shelf, the encoder design is a drop-in component: any method that consumes point features can adopt the stratification, and the paper suggests combining with object-compositional methods as future work.

Reading between the lines

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

  • If the frequency-specialization story is right, the same stratification should transfer to radiance-field-only models and to multi-resolution grid encoders, where frequency bands are already separated spatially rather than by octave; a natural test is whether per-octave feature decorrelation improves those baselines as well.
  • The weighting module is an implicit diversity regularizer: by discouraging redundancy among feature columns, it resembles decorrelation penalties, and one could push further by adding an explicit loss that penalizes mutual information between band features, then check whether the gains grow.
  • The paper's evidence for geometric improvement is indirect, so the clean way to confirm the central claim is to measure Chamfer distance and F-score against ground-truth meshes on a dataset that provides them; if those metrics do not improve with rendering, the claim would need to be narrowed to rendering fidelity.
  • The demonstrated robustness to N=12 suggests a practical recipe: raise the number of frequency levels once stratification is in place, since the low-frequency encoder is protected from the instability that high octaves introduce in single-encoder training.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper proposes FreBIS, a modification of neural implicit surface reconstruction in which the single positional-encoding encoder of VolSDF is replaced by three parallel encoders, each dedicated to a band of positional frequencies (low, middle, high). A redundancy-aware weighting module computes feature importance from pairwise similarities and uses these weights to combine the three encodings before decoding into an SDF value and appearance feature. The model is trained with photometric and Eikonal losses and evaluated on 9 BlendedMVS scenes against VolSDF and a parameter-matched scaled-up VolSDF, reporting PSNR, SSIM, and LPIPS for rendering plus qualitative mesh comparisons.

Significance. If the claims are supported, FreBIS offers a simple and potentially generalizable architectural improvement for neural implicit surface methods: frequency stratification across multiple encoders, with a matched-parameter baseline to control for capacity. The inclusion of the Scaled-up VolSDF baseline is a genuine strength, as is the per-scene reporting across 9 scenes. The gains over baselines are consistent but small, and the central surface-reconstruction claim currently rests on qualitative mesh comparisons rather than quantitative geometry metrics, so the significance of the paper depends on additional validation.

major comments (4)
  1. [§5.1, §5.2, Table 1] The claim of improved reconstructed 3D surfaces is not supported by quantitative evidence. Section 5.1 states that ground-truth meshes are unavailable for BlendedMVS, and Table 1 reports only rendering metrics (PSNR, SSIM, LPIPS); the mesh-quality comparison is qualitative (Fig. 5). Rendering fidelity does not directly measure geometric accuracy, so the abstract's claim of improvements in the 'quality of the reconstructed 3D surfaces' goes beyond the presented evidence. Please add a quantitative geometry evaluation on a dataset with ground-truth meshes (e.g., DTU, using Chamfer distance or F-score), or restrict the claim to rendering fidelity.
  2. [§5.2, Table 1] The reported improvements over baselines are small (mean PSNR 27.38 vs 27.23 for VolSDF and 26.90 for Scaled-up VolSDF; LPIPS 0.024 vs 0.025 and 0.027) and are given as single runs without error bars, number of seeds, or any significance test. The word 'significant' in the abstract is therefore not established. Please provide variance estimates across multiple runs or rephrase the claim to avoid overstatement.
  3. [§4.1, §5.2, Table 1] The design does not isolate the effect of the frequency split as the cause of the gains. The comparison against Scaled-up VolSDF changes both parameter count and architecture, and no ablation is reported in which a three-encoder architecture receives the full set of positional frequencies in every branch. Without such a control, the improvements could stem from the multi-branch ensemble rather than from frequency stratification. Please add an ablation with identical three-encoder architecture where all encoders get the same full positional encoding.
  4. [§4.2, Eqs. (4)–(7), Table 3] The paper states that the redundancy-aware weighting module 'promotes mutual dissimilarity' of the encoded features, but it does not contain a loss term enforcing dissimilarity; it only reassigns weights based on the current similarity of the features. No metric of feature redundancy or complementarity is reported, and the quantitative effect of the module on the Dog scene is a small gain (+0.25 dB PSNR over simple averaging). The band-wise mesh visualizations in Fig. 7 are expected given that each encoder receives different frequency inputs. Please provide a quantitative redundancy measure (e.g., average pairwise absolute cosine similarity of encoder features) or soften the claim about promoting dissimilarity.
minor comments (4)
  1. [§5.1] The text says 'The training loss (Eq. 4) is computed with λ = 0.1', but Eq. 4 is the similarity matrix in the redundancy module; the training loss is Eq. 8. Please correct the cross-reference.
  2. [Table 2] Table 2 is difficult to read because many entries appear concatenated without spaces (e.g., '26.22 27.4827.2930.52'). Please reformat the table with proper cell boundaries.
  3. [Supplementary Table 6] In the supplementary, Table 6 uses the header 'NL,NM,NH' for what is described as a study of the number of encoder layers; since NL,NM,NH was previously used for frequency counts, please use distinct notation (e.g., LL, LM, LH) to avoid confusion.
  4. [§5.2] The phrase 'gains of up to 2% on SSIM' is ambiguous about whether the percentage is relative or absolute; please specify the computation.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the central claim is an empirical comparison against external baselines and data, not a reduction to the paper's own definitions or fitted parameters.

full rationale

The paper's central claim is that replacing the standard encoder in VolSDF with the proposed frequency-stratified encoders improves reconstruction and rendering; this is evaluated on BlendedMVS against external photometric metrics (PSNR, SSIM, LPIPS) and qualitative mesh inspection. The redundancy-aware weighting module does compute weights from the features it then combines (Eqs. 4-7), but this is a self-referential architectural mechanism rather than a circular derivation: the claimed improvement is not defined in terms of those weights, and the ablation compares against simple averaging using the same external rendering metrics. No parameter is fitted to the evaluation target and then reported as a prediction. No load-bearing self-citation appears: VolSDF is cited as an external baseline and decoder backbone, not as justification for FreBIS's validity. No uniqueness theorem or prior author result is invoked to force the design choice. The main weakness, that surface quality is assessed only qualitatively because ground-truth meshes are unavailable for BlendedMVS, is an evidence and correctness concern rather than a circularity. The derivation and evaluation are self-contained relative to external image data, so the circularity score is 0.

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

The paper introduces no new physical entities, forces, or conserved quantities. The free parameters listed are hand-set hyperparameters; the axioms are standard assumptions in neural implicit surface reconstruction, with the frequency-split choice being specific to this paper and tuned on the evaluation scenes.

free parameters (3)
  • softmax temperature tau = 0.5
    Set by hand in Sec. 4.2, controls weight sharpness, not ablated.
  • Eikonal loss weight lambda = 0.1
    Set by hand in Sec. 5.1, not ablated.
  • Number of frequency levels N = 6
    Chosen in Sec. 4.1; supplementary shows N=9 and N=12 give similar mean PSNR (27.36 and 27.21), so the result is not very sensitive to this choice.
assumptions (4)
  • domain assumption Photometric loss plus Eikonal regularization is sufficient for SDF learning without ground-truth geometry.
    Borrowed from VolSDF and NeuS; the paper relies on this without re-validation in Sec. 4.4.
  • domain assumption The 9 BlendedMVS scenes are representative and the per-scene results are stable under small architecture changes.
    Section 5.1 selects scenes following prior work; no uncertainty or repeated-run analysis is provided, and the Bread scene sometimes regresses.
  • ad hoc to paper Equal split of N=6 positional frequencies across three encoders covers the scene frequency range.
    Chosen in Sec. 4.1 and Sec. 5.1; the ablation in Table 2 shows even split is most stable, but the choice is tuned on the evaluation scenes.
  • domain assumption The off-the-shelf VolSDF decoder is a strong enough backbone that the encoder change isolates the benefit.
    Section 4.3 uses the VolSDF decoder; a scaled-up VolSDF baseline is added, but no other decoders are tested.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FreBIS: Frequency-Based Stratification for Neural Implicit Surface Representations." pith.science (2026). https://pith.science/paper/54KZOIKA

@misc{pith2026250420222,
  author       = {Pith},
  title        = {Pith review of: FreBIS: Frequency-Based Stratification for Neural Implicit Surface Representations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/54KZOIKA}},
  note         = {Machine review of arXiv:2504.20222}
}
read the original abstract

Neural implicit surface representation techniques are in high demand for advancing technologies in augmented reality/virtual reality, digital twins, autonomous navigation, and many other fields. With their ability to model object surfaces in a scene as a continuous function, such techniques have made remarkable strides recently, especially over classical 3D surface reconstruction methods, such as those that use voxels or point clouds. However, these methods struggle with scenes that have varied and complex surfaces principally because they model any given scene with a single encoder network that is tasked to capture all of low through high-surface frequency information in the scene simultaneously. In this work, we propose a novel, neural implicit surface representation approach called FreBIS to overcome this challenge. FreBIS works by stratifying the scene based on the frequency of surfaces into multiple frequency levels, with each level (or a group of levels) encoded by a dedicated encoder. Moreover, FreBIS encourages these encoders to capture complementary information by promoting mutual dissimilarity of the encoded features via a novel, redundancy-aware weighting module. Empirical evaluations on the challenging BlendedMVS dataset indicate that replacing the standard encoder in an off-the-shelf neural surface reconstruction method with our frequency-stratified encoders yields significant improvements. These enhancements are evident both in the quality of the reconstructed 3D surfaces and in the fidelity of their renderings from any viewpoint.

Figures

Figures reproduced from arXiv: 2504.20222 by the authors.

Figure 1
Figure 1. Overview of FreBIS: (a) Frequency-domain Represen￾tation: FreBIS works by mapping the input point coordinate to the frequency domain and encoding it via three frequency-band encoders – one each for low, middle, and high. (b) Redundancy￾aware Weighting: This module computes weights that indicate the importance of the three encoded features according to the dissim￾ilarity of each to the other two. These weights are th… view at source ↗
Figure 2
Figure 2. FreBIS framework: Given an input 3D point [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Redundancy-aware weighting module: The redundancy-aware weighting module takes the encoded frequency features and [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of viewpoint-based scene rendering on the BlendedMVS dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparison of surface reconstruction quality for the BlendedMVS dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Visualization of norms of weighted feature vectors, [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Reconstructed meshes for each frequency band. [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Qualitative comparison on the capability to deal with higher frequencies. [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]
Figure 9
Figure 9. Figure 9: Qualitative comparison of viewpoint-based scene rendering with varying number of frequencies. [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: Qualitative comparison on surface reconstruction with a different number of frequencies. [PITH_FULL_IMAGE:figures/full_fig_p016_10.png]
Figure 11
Figure 11. Figure 11: Qualitative comparison on viewpoint-based scene rendering using FreBIS, obtained by varying the number of encoder layers. [PITH_FULL_IMAGE:figures/full_fig_p017_11.png]
Figure 12
Figure 12. Figure 12: Qualitative comparison based on 3D surface reconstruction using FreBIS, obtained by varying the number of encoder layers. [PITH_FULL_IMAGE:figures/full_fig_p017_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 44 canonical work pages

  1. [1]

    Neural RGB-D surface reconstruction, 2022

    Dejan Azinovi ´c, Ricardo Martin-Brualla, Dan B Goldman, Matthias Nießner, and Justus Thies. Neural RGB-D surface reconstruction, 2022. 2

  2. [2]

    Patchmatch: a randomized correspondence algorithm for structural image editing

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

  3. [3]

    De Bonet

    Jeremy S. De Bonet. Poxels: Probabilistic voxelized vol- ume reconstruction. In Proceedings of the IEEE Interna- tional Conference on Computer Vision (ICCV), 1999. 2

  4. [4]

    Broadhurst, T.W

    A. Broadhurst, T.W. Drummond, and R. Cipolla. A proba- bilistic framework for space carving. InProceedings of IEEE International Conference on Computer Vision , pages 388– 393 vol.1, 2001. 2

  5. [5]

    Mobilenerf: Exploiting the polygon ras- terization pipeline for efficient neural field rendering on mo- bile architectures

    Zhiqin Chen, Thomas Funkhouser, Peter Hedman, and An- drea Tagliasacchi. Mobilenerf: Exploiting the polygon ras- terization pipeline for efficient neural field rendering on mo- bile architectures. In Proceedings of The Conference on Computer Vision and Pattern Recognition (CVPR), 2023. 3

  6. [6]

    High-quality surface re- construction using gaussian surfels

    Pinxuan Dai, Jiamin Xu, Wenxiang Xie, Xinguo Liu, Huamin Wang, and Weiwei Xu. High-quality surface re- construction using gaussian surfels. In Proceedings of ACM SIGGRAPH 2024 Conference Papers. Association for Com- puting Machinery, 2024. 3

  7. [7]

    Improving neural im- plicit surfaces geometry with patch warping

    Franc ¸ois Darmon, B´en´edicte Bascle, Jean-Cl ´ement Devaux, Pascal Monasse, and Mathieu Aubry. Improving neural im- plicit surfaces geometry with patch warping. In Proceedings of The Conference on Computer Vision and Pattern Recogni- tion (CVPR), 2022. 2

  8. [8]

    Faugeras and R

    O. Faugeras and R. Keriven. Variational principles, sur- face evolution, pdes, level set methods, and the stereo prob- lem. IEEE Transactions on Image Processing, 7(3):336–344,

Show all 62 references
  1. [9]

    Faugeras and Renaud Keriven

    Olivier D. Faugeras and Renaud Keriven. Complete dense stereovision using level set methods. In Proceedings of the 5th European Conference on Computer Vision-Volume I - Volume I, page 379–393, Berlin, Heidelberg, 1998. Springer- Verlag. 2

  2. [10]

    Geo-Neus: Geometry-consistent neural implicit sur- faces learning for multi-view reconstruction, 2022

    Qiancheng Fu, Qingshan Xu, Yew-Soon Ong, and Wenbing Tao. Geo-Neus: Geometry-consistent neural implicit sur- faces learning for multi-view reconstruction, 2022. 2

  3. [11]

    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, 2010. 2

  4. [12]

    Massively parallel multiview stereopsis by surface normal diffusion

    Silvano Galliani, Katrin Lasinger, and Konrad Schindler. Massively parallel multiview stereopsis by surface normal diffusion. In Proceedings of the IEEE International Confer- ence on Computer Vision (ICCV), pages 873–881, 2015. 2

  5. [13]

    Michael Goesele, Noah Snavely, Brian Curless, Hugues Hoppe, and Steven M. Seitz. Multi-view stereo for commu- nity photo collections. In Proceedings of IEEE 11th Inter- national Conference on Computer Vision , pages 1–8, 2007. 2

  6. [14]

    HIVE: HIerarchical V olume Encoding for Neu- ral Implicit Surface Reconstruction, 2024

    Xiaodong Gu, Weihao Yuan, Heng Li, Zilong Dong, and Ping Tan. HIVE: HIerarchical V olume Encoding for Neu- ral Implicit Surface Reconstruction, 2024. arXiv:2408.01677 [cs]. 2

  7. [15]

    Neural 3D Scene Reconstruction from Multiple 2D Images without 3D Supervision, 2023

    Yi Guo, Che Sun, Yunde Jia, and Yuwei Wu. Neural 3D Scene Reconstruction from Multiple 2D Images without 3D Supervision, 2023. arXiv:2306.17643 [cs]. 2

  8. [16]

    SuGaR: Surface- aligned gaussian splatting for efficient 3d mesh reconstruc- tion and high-quality mesh rendering

    Antoine Gu ´edon and Vincent Lepetit. SuGaR: Surface- aligned gaussian splatting for efficient 3d mesh reconstruc- tion and high-quality mesh rendering. In Proceedings of The Conference on Computer Vision and Pattern Recogni- tion (CVPR). arXiv, 2023. arXiv:2311.12775 [cs]. 3

  9. [17]

    Hierarchical volu- metric multi-view stereo reconstruction of manifold surfaces based on dual graph embedding

    Alexander Hornung and Leif Kobbelt. Hierarchical volu- metric multi-view stereo reconstruction of manifold surfaces based on dual graph embedding. In Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1, page 503–510, U...

  10. [18]

    2d gaussian splatting for geometrically ac- curate radiance fields

    Binbin Huang, Zehao Yu, Anpei Chen, Andreas Geiger, and Shenghua Gao. 2d gaussian splatting for geometrically ac- curate radiance fields. In Proceedings of SIGGRAPH 2024 Conference Papers. Association for Computing Machinery,

  11. [19]

    3d gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Transactions on Graphics, 42 (4), 2023. 3

  12. [20]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization, 2017. 6

  13. [21]

    Kutulakos and Steven M

    Kiriakos N. Kutulakos and Steven M. Seitz. A theory of shape by space carving. Proceedings of the Seventh IEEE In- ternational Conference on Computer Vision, 38(3):199–218,

  14. [22]

    Efficient multi-view reconstruction of large-scale scenes us- ing interest points, delaunay triangulation and graph cuts

    Patrick Labatut, Jean-Philippe Pons, and Renaud Keriven. Efficient multi-view reconstruction of large-scale scenes us- ing interest points, delaunay triangulation and graph cuts. In Proceedings of IEEE 11th International Conference on Com- puter Vision, pages 1–8, 2007. 2

  15. [23]

    V ox-surf: V oxel-based implicit sur- face representation

    Hai Li, Xingrui Yang, Hongjia Zhai, Yuqian Liu, Hujun Bao, and Guofeng Zhang. V ox-surf: V oxel-based implicit sur- face representation. IEEE Transactions on Visualization and Computer Graphics, 30(3):1743–1755, 2024. 3

  16. [24]

    Splatsdf: Boosting neural implicit sdf via gaussian splatting fusion,

    Runfa Blark Li, Keito Suzuki, Bang Du, Ki Myung Brian Lee, Nikolay Atanasov, and Truong Nguyen. Splatsdf: Boosting neural implicit sdf via gaussian splatting fusion,

  17. [25]

    RICO: Regularizing the Unobservable for Indoor Compositional Reconstruction

    Zizhang Li, Xiaoyang Lyu, Yuanyuan Ding, Mengmeng Wang, Yiyi Liao, and Yong Liu. RICO: Regularizing the Unobservable for Indoor Compositional Reconstruction. In Proceedings of the IEEE International Conference on Com- puter Vision (ICCV). arXiv, 2023. arXiv:2303.08605 [cs]. 3, 8

  18. [26]

    Taylor, Mathias Unberath, Ming-Yu Liu, and Chen-Hsuan Lin

    Zhaoshuo Li, Thomas M ¨uller, Alex Evans, Russell H. Taylor, Mathias Unberath, Ming-Yu Liu, and Chen-Hsuan Lin. Neuralangelo: High-Fidelity Neural Surface Recon- struction. In Proceedings of IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR) . arXiv, 2023. arX...

  19. [27]

    HelixSurf: A robust and efficient neural implicit sur- face learning of indoor scenes with iterative intertwined reg- ularization

    Zhihao Liang, Zhangjin Huang, Changxing Ding, and Kui Jia. HelixSurf: A robust and efficient neural implicit sur- face learning of indoor scenes with iterative intertwined reg- ularization. In Proceedings of IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR) ....

  20. [28]

    3DGSR: Implicit surface reconstruction with 3d gaussian splatting, 2024

    Xiaoyang Lyu, Yang-Tian Sun, Yi-Hua Huang, Xiuzhe Wu, Ziyi Yang, Yilun Chen, Jiangmiao Pang, and Xiaojuan Qi. 3DGSR: Implicit surface reconstruction with 3d gaussian splatting, 2024. 3

  21. [29]

    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 Proceedings of The European Conference on Com- puter Vision (ECCV), 2020. 2, 3

  22. [30]

    Differentiable V olumetric Render- ing: Learning Implicit 3D Representations without 3D Su- pervision

    Michael Niemeyer, Lars Mescheder, Michael Oechsle, and Andreas Geiger. Differentiable V olumetric Render- ing: Learning Implicit 3D Representations without 3D Su- pervision. In Proceedings of The Conference on Com- puter Vision and Pattern Recognition (CVPR) . arXiv, 2020. arX...

  23. [31]

    UNISURF: Unifying neural implicit surfaces and radi- ance fields for multi-view reconstruction

    Michael Oechsle, Songyou Peng, and Andreas Geiger. UNISURF: Unifying neural implicit surfaces and radi- ance fields for multi-view reconstruction. In Interna- tional Conference on Computer Vision (ICCV). arXiv, 2021. arXiv:2104.10078 [cs]. 2

  24. [32]

    Quasi-dense similarity learn- ing for multiple object tracking

    Jiangmiao Pang, Linlu Qiu, Xia Li, Haofeng Chen, Qi Li, Trevor Darrell, and Fisher Yu. Quasi-dense similarity learn- ing for multiple object tracking. InProceeding of the Confer- ence on Computer Vision and Pattern Recognition (CVPR) ,

  25. [33]

    H2O- SDF: Two-phase Learning for 3D Indoor Reconstruction us- ing Object Surface Fields

    Minyoung Park, Mirae Do, YeonJae Shin, Jaeseok Yoo, Jongkwang Hong, Joongrock Kim, and Chul Lee. H2O- SDF: Two-phase Learning for 3D Indoor Reconstruction us- ing Object Surface Fields. In Proceedings of The Interna- tional Conference on Learning Representations. arXiv, 2024. ...

  26. [34]

    Pytorch: An im- perative style, high-performance deep learning library, 2019

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas K ¨opf, Edward Yang, Zach DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu F...

  27. [35]

    Normal- guided Detail-Preserving Neural Implicit Functions for High-Fidelity 3D Surface Reconstruction, 2024

    Aarya Patel, Hamid Laga, and Ojaswa Sharma. Normal- guided Detail-Preserving Neural Implicit Functions for High-Fidelity 3D Surface Reconstruction, 2024. arXiv:2406.04861 [cs]. 2

  28. [36]

    NeRFMeshing: Distilling Neural Radi- ance Fields into Geometrically-Accurate 3D Meshes, 2023

    Marie-Julie Rakotosaona, Fabian Manhardt, Diego Mar- tin Arroyo, Michael Niemeyer, Abhijit Kundu, and Fed- erico Tombari. NeRFMeshing: Distilling Neural Radi- ance Fields into Geometrically-Accurate 3D Meshes, 2023. arXiv:2303.09431 [cs]. 3

  29. [37]

    Sch ¨onberger, Enliang Zheng, Jan-Michael Frahm, and Marc Pollefeys

    Johannes L. Sch ¨onberger, Enliang Zheng, Jan-Michael Frahm, and Marc Pollefeys. Pixelwise view selection for un- structured multi-view stereo. In Proceedings of European Conference on Computer Vision , pages 501–518, Cham,

  30. [38]

    Seitz and C.R

    S.M. Seitz and C.R. Dyer. Photorealistic scene reconstruc- tion by voxel coloring. In Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recog- nition (CVPR), pages 1067–1073, 1997. 2

  31. [39]

    Delicate textured mesh recovery from nerf via adaptive surface refinement

    Jiaxiang Tang, Hang Zhou, Xiaokang Chen, Tianshu Hu, Er- rui Ding, Jingdong Wang, and Gang Zeng. Delicate textured mesh recovery from nerf via adaptive surface refinement. In Proceedings of the IEEE International Conference on Com- puter Vision (ICCV), 2023. 3

  32. [40]

    High accuracy and visibility-consistent dense multiview stereo

    Hoang-Hiep Vu, Patrick Labatut, Jean-Philippe Pons, and Renaud Keriven. High accuracy and visibility-consistent dense multiview stereo. IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 34(5):889–901, 2012. 2

  33. [41]

    GO- Surf: Neural Feature Grid Optimization for Fast, High- Fidelity RGB-D Surface Reconstruction

    Jingwen Wang, Tymoteusz Bleja, and Lourdes Agapito. GO- Surf: Neural Feature Grid Optimization for Fast, High- Fidelity RGB-D Surface Reconstruction. In Proceedings of International Conference on 3D Vision (3DV) . arXiv, 2022. arXiv:2206.14735 [cs]. 2

  34. [42]

    NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal Priors

    Jiepeng Wang, Peng Wang, Xiaoxiao Long, Christian Theobalt, Taku Komura, Lingjie Liu, and Wenping Wang. NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal Priors. In Proceedings of European Conference on Computer Vision. arXiv, 2022. arXiv:2206.13597 [cs]. 2

  35. [43]

    NeuS: Learning Neu- ral Implicit Surfaces by V olume Rendering for Multi- view Reconstruction

    Peng Wang, Lingjie Liu, Yuan Liu, Christian Theobalt, Taku Komura, and Wenping Wang. NeuS: Learning Neu- ral Implicit Surfaces by V olume Rendering for Multi- view Reconstruction. In Proceedings of 35th Conference on Neural Information Processing Systems . arXiv, 2021. arXiv:2...

  36. [44]

    HF- NeuS: Improved Surface Reconstruction Using High- Frequency Details

    Yiqun Wang, Ivan Skorokhodov, and Peter Wonka. HF- NeuS: Improved Surface Reconstruction Using High- Frequency Details. In Proceedings of the 36th Interna- tional Conference on Neural Information Processing Sys- tems. arXiv, 2022. arXiv:2206.07850 [cs]. 2

  37. [45]

    NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Recon- struction, 2023

    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, 2023. arXiv:2212.05231 [cs]. 3

  38. [46]

    PET- NeuS: Positional Encoding Tri-Planes for Neural Sur- faces

    Yiqun Wang, Ivan Skorokhodov, and Peter Wonka. PET- NeuS: Positional Encoding Tri-Planes for Neural Sur- faces. In Proceedings of The Conference on Com- puter Vision and Pattern Recognition (CVPR) . arXiv, 2023. arXiv:2305.05594 [cs]. 2

  39. [47]

    Bovik, H.R

    Zhou Wang, A.C. Bovik, H.R. Sheikh, and E.P. Simoncelli. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing , 13(4): 600–612, 2004. 6

  40. [48]

    Object- Compositional Neural Implicit Surfaces

    Qianyi Wu, Xian Liu, Yuedong Chen, Kejie Li, Chuanxia Zheng, Jianfei Cai, and Jianmin Zheng. Object- Compositional Neural Implicit Surfaces. In Proceedings of European Conference on Computer Vision . arXiv, 2022. arXiv:2207.09686 [cs]. 3, 8

  41. [49]

    ObjectSDF++: Improved object-compositional neural implicit surfaces

    Qianyi Wu, Kaisiyuan Wang, Kejie Li, Jianmin Zheng, and Jianfei Cai. ObjectSDF++: Improved object-compositional neural implicit surfaces. In Proceedings of the IEEE/CVF International Conference on Computer Vision. arXiv, 2023. arXiv:2308.07868 [cs]. 3

  42. [50]

    Blended- MVS: A large-scale dataset for generalized multi-view stereo networks

    Yao Yao, Zixin Luo, Shiwei Li, Jingyang Zhang, Yufan Ren, Lei Zhou, Tian Fang, and Long Quan. Blended- MVS: A large-scale dataset for generalized multi-view stereo networks. In Proceedings of The Conference on Com- puter Vision and Pattern Recognition (CVPR) . arXiv, 2020. arX...

  43. [51]

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

    Lior Yariv, Yoni Kasten, Dror Moran, Meirav Galun, Matan Atzmon, Ronen Basri, and Yaron Lipman. Multiview neu- ral surface reconstruction by disentangling geometry and ap- pearance. Advances in Neural Information Processing Sys- tems, 2020. arXiv:2003.09852 [cs]. 2

  44. [52]

    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. In Proceedings of The Conference on Neural Information Processing Sys- tems. arXiv, 2021. arXiv:2106.12052 [cs]. 2, 3, 4, 6, 1

  45. [53]

    Srinivasan, Richard Szeliski, Jonathan T

    Lior Yariv, Peter Hedman, Christian Reiser, Dor Verbin, Pratul P. Srinivasan, Richard Szeliski, Jonathan T. Barron, and Ben Mildenhall. Bakedsdf: Meshing neural sdfs for real- time view synthesis, 2023. 3

  46. [54]

    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 (NeurIPS), 2022. arXiv:2206.00665 [cs]. 2

  47. [55]

    Efros, Eli Shecht- man, and Oliver Wang

    Richard Zhang, Phillip Isola, Alexei A. Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of The Conference on Computer Vision and Pattern Recogni- tion (CVPR), 2018. 6 FreBIS: Frequency-Based Strati...

  48. [57]

    Ablation study of the redundancy-aware weightingmod- ule

  49. [58]

    Comparative study of the number of frequency levels

  50. [59]

    Comparative study of encoder architecture variants

  51. [60]

    Table 4 shows quantitative comparison results of the FreBIS with and without this module

    Ablation study of the redundancy-aware weighting module A key innovation of FreBIS is the rendundancy-aware weighting module which combines the complementary in- formation from the different encoders by promoting mutual dissimilarity. Table 4 shows quantitative comparison resu...

  52. [61]

    As shown in Table 5 and Fig

    Comparative study of the number of fre- quency levels We conduct experiments to study the effect of the choice of frequency levels N for both FreBIS and Scaled-up V olSDF [52]. As shown in Table 5 and Fig. 8, the Scaled- up V olSDF is sensitive to the choice of frequency level...

  53. [62]

    As seen from the results in Table 6 as well as Fig

    Comparative study of encoder architecture variants In order to design the encoders of FreBIS optimally, we study the effect of varying the number of layers of each of the three encoders of FreBIS and compare their perfor- mances. As seen from the results in Table 6 as well as ...

  54. [2016]

    Springer International Publishing. 2

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.