Pith. sign in

REVIEW 4 major objections 6 minor 56 references

Point upsampling networks for single-photon sensing

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper proposes treating single-photon point cloud reconstruction as a point upsampling task, and reports that a Mamba-based network, SPU-MAMBA, achieves best or near-best accuracy under distortion noise.

desk verdict A solid point-upsampling paper with a new application to single-photon clouds, but the single-photon relevance rests on an unvalidated Gaussian noise model and the 'first' claim is asserted, not shown. read the letter →

arxiv 2508.12986 v1 pith:DP34RZAS submitted 2025-08-18 physics.optics cs.CV

Jinyi Liu , Guoyang Zhao , Lijun Liu , Yiguang Hong , Weiping Zhang , Shuming Cheng This is my paper · ORCID
classification physics.opticscs.CV
keywords single-photonsensingpointcloudupsamplingstatespacemodelMambadepthdistortionLiDARdensificationnoiserobustness
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

Single-photon sensing yields point clouds that are sparse and spatially biased, which limits their practical use in 3D tasks. This paper argues that the problem can be framed as point cloud upsampling: a neural network takes the sparse, distorted cloud and generates additional points on the underlying surface while correcting depth offset. The proposed network, SPU-MAMBA, uses a state-space (Mamba) backbone with multi-path scanning and an adaptive shift module, achieving linear computational complexity. On standard benchmarks with added depth-offset noise, it reports the best or second-best Chamfer distance, Hausdorff distance, and point-to-surface error among eight methods, and shows flatter degradation as noise increases. If correct, this offers a direct post-processing route to make single-photon imaging outputs denser and more geometrically faithful for downstream tasks.

What carries the argument

The load-bearing machinery is the SPU-MAMBA network architecture, organized into a point encoder, a Mamba decoder, and an upsample shift module. The decoder's multi-path scanning mechanism (six axis-ordered sortings such as XYZ, XZY, YXZ, and their permutations) serializes a point cloud into permutation-equivariant sequences, preserving spatial coherence and depth-layer continuity. Bidirectional Mamba blocks — a state-space model with forward and backward passes plus convolutional local-feature extraction — process each sequence in linear time, suppressing noise propagation while capturing long-range geometry. The upsample shift module uses a two-stage deformable kernel-point convolution to predict per-point displacements for $r_c$ generated points, with a repulsion loss that encourages uniform distribution. The key combination is that the scanning order captures long-range geometry, the bidirectional SSM balances global and local detail, and the deformable shift corrects offset-induced distortion, together producing the reported accuracy and robustness.

What would settle it

Evaluate SPU-MAMBA and the baseline methods on real single-photon LiDAR scenes with known ground-truth geometry (for example, a calibrated target scene scanned at multiple ranges), measure the same CD, HD, and P2F metrics against the ground truth, and check whether SPU-MAMBA still holds its ranking. As a quicker check, measure the actual distribution of depth errors in a real SPAD point cloud: if the error is non-Gaussian, has a non-zero mean, or is strongly correlated with surface reflectivity or incidence angle, then the paper's noise-augmentation procedure does not capture the true distortion and the reported robustness may not hold outside the synthetic setup.

Watch

Extended reading notes

Core claim

The central claim is that point upsampling (PU) networks can serve as a general post-processing tool for single-photon sensing: given a sparse, distorted point cloud, a PU network can increase point density by a factor $r_c$ and simultaneously reduce offset-induced distortion, yielding a more uniform and geometrically accurate representation. The specific instantiation, SPU-MAMBA, integrates a multi-path scanning mechanism that serializes points along six axis-ordered paths, a bidirectional Mamba (state-space) backbone that captures global context and local detail with linear complexity, and an upsample shift module based on deformable kernel-point convolution that displaces generated points to fit local geometry. On the PU-GAN and PU1K benchmarks with additive Gaussian depth offsets of $\sigma$ up to 0.1, the authors report that SPU-MAMBA scores the best Chamfer distance at most noise levels, consistently second-best Hausdorff distance, and top or near-top point-to-surface error, while degrading more gracefully than baselines. On real-world single-photon LiDAR data, it is claimed to generate visually consistent, detail-preserving, and noise-suppressed point clouds.

Load-bearing premise

The synthetic zero-mean Gaussian depth offset added to every point is assumed to reproduce the real bias structure of single-photon point clouds, which actually arises from the correlation between photon arrival time and emission angle and from atmospheric scattering; if it does not, the reported robustness gains will not transfer to real single-photon sensing.

Editorial extensions

If this is right

  • SPU-MAMBA can upsample non-uniformly distributed point clouds with linear computational complexity, making the approach scalable to large-scale scenes where quadratic attention would be prohibitive.
  • A single model, when trained on noise-augmented point clouds, maintains more stable Chamfer distance and point-to-surface error as depth-offset noise increases from $\sigma=0$ to $\sigma=0.1$, degrading more gracefully than the eight baselines.
  • The framework can be applied directly to real-world single-photon LiDAR data without ground truth, producing point clouds that are visually more uniform, detail-preserving, and resilient to residual noise and outliers.
  • Because the upsampled clouds better preserve local geometry and reduce spatial distortion, downstream tasks such as object detection, segmentation, and geometric classification stand to receive higher-quality input point clouds.
  • The authors position this as the first upsampling framework for single-photon sensing, opening a distinct processing stage that complements model-based denoising and learning-based reconstruction.

Reading between the lines

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

  • If the synthetic zero-mean Gaussian depth-offset model matches the real bias structure, the same network could be trained entirely on simulated LiDAR histograms, since the loss functions only require the noisy input and a clean surface model — no paired real-world ground truth needed.
  • The multi-path scanning mechanism may transfer to other modalities that produce sparse, distorted point clouds, such as underwater sonar or long-range radar, where similar reflectivity-driven sampling bias occurs.
  • A testable extension is to quantify whether the upsampled point clouds actually improve downstream task accuracy (for example, object detection mAP or segmentation IoU), which the paper motivates but does not measure directly.
  • The robustness claim is demonstrated only for zero-mean Gaussian depth offsets; real distortions are likely correlated with reflectivity and incidence angle, so a stress test with a physics-based noise model would clarify the practical limits of the approach.
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 / 6 minor

Summary. The paper proposes SPU-MAMBA, a point upsampling network for single-photon point clouds, built on a Mamba/state-space backbone with a multi-path scanning mechanism, a bidirectional Mamba decoder, and an adaptive upsample shift module. The authors evaluate SPU-MAMBA on the PU-GAN and PU1K benchmarks under synthetic Gaussian depth-offset noise at four levels (σ = 0, 0.02, 0.06, 0.1), comparing with seven generic point upsampling baselines using Chamfer distance, Hausdorff distance, and point-to-surface distance. A qualitative real-world single-photon LiDAR experiment is also reported. The paper claims best or second-best performance across most metrics and noise levels, and states that it is the first to establish the upsampling framework for single-photon sensing.

Significance. If the robustness claims hold, the work would be a useful application of point upsampling to a domain where point clouds are sparse and distorted, and the architectural components are assembled coherently. The paper's strength is its systematic comparison across eight methods and four noise levels on two benchmarks, with internally consistent tables. However, the significance is currently limited by three load-bearing issues: the synthetic Gaussian noise model is not validated against real single-photon distortion, the real-world evaluation is only qualitative, and the point-to-surface metric as defined in Eq. (13) does not measure distance to any ground-truth surface. The absence of single-photon-specific baselines further weakens the domain-novelty claim.

major comments (4)
  1. [Sec. IV / Sec. V.D] The noise model is unvalidated. Section IV states that a random depth offset sampled from a zero-mean Gaussian with σ ∈ [0,0.1] is added to every point 'to account for the correlation between photon arrival time and emission angle, and the atmospheric scattering effect.' Real single-photon depth errors are not generally zero-mean Gaussian; they include systematic, scene-dependent biases from photon arrival statistics, incidence angle, and atmospheric scattering. The real-world evaluation in Sec. V.D is purely qualitative, so the paper does not demonstrate that the synthetic corruption reproduces the structure of real single-photon point-cloud distortion. Please validate the noise model against real data or provide a quantitative mapping of σ to physical units and real-system parameters.
  2. [Sec. V.A / Tables I-II] The reported performance differences are not accompanied by error bars, confidence intervals, or significance tests. Some claimed improvements are very small; for example, in Table II at σ = 0, SPU-MAMBA's CD is 0.485 × 10⁻³, which is worse than PU-GCN's 0.479 × 10⁻³, and at σ = 0.06 and 0.1 the CD gaps are only 1.8% and 6.5%. Without repeated runs and variance estimates, the central claims of 'best/second-best' performance and 'stronger noise robustness' are not statistically supported.
  3. [Eq. (13) / Tables I-II] The definition of P2F is ambiguous. The text identifies the point-to-surface distance P2F with Eq. (13), but Eq. (13) as written measures the distance between deformed kernel points and the input neighborhood N(p), not the distance to any ground-truth surface. A model that keeps output points near the input can minimize this quantity without improving reconstruction fidelity. If the reported P2F is instead computed against a ground-truth mesh or surface, that metric must be defined separately and reported; as written, the P2F improvements in Tables I and II are not a meaningful fidelity claim.
  4. [Sec. II.B / Sec. V] The novelty claim is not backed by the comparison set. The abstract and introduction state that this is the first upsampling framework for single-photon sensing, but all baselines are generic point upsampling networks; none of the single-photon-specific reconstruction methods reviewed in Sec. II.B (e.g., first-photon imaging [1], non-local neural network [40], U-Net++ [41], or pixel-wise residual shrinkage net [42]) is included in the experiments. To substantiate the domain-specific contribution, at least one single-photon reconstruction or denoising method should be adapted as a baseline, or the novelty claim should be tempered.
minor comments (6)
  1. [Eqs. (13)-(14) / Sec. IV] The symbol σ is used both for the noise standard deviation in Sec. IV and for the distance scaling factor in Eqs. (13) and (14); please use different symbols to avoid ambiguity.
  2. [Sec. V.C] Please state explicitly that the synthetic noise is applied only to the input point clouds and not to the ground-truth point clouds; the text 'added to every point in the dataset' is ambiguous.
  3. [Sec. II.C] Appendix A is referenced in Sec. II.C but does not appear in the manuscript; please include it or remove the reference.
  4. [Sec. VII] In the Conclusions, 'sing-photon point cloud' should be corrected to 'single-photon point cloud'.
  5. [Sec. III] The claim of linear computational complexity should be qualified: the multi-path scanning and the K-nearest-neighbor construction in Sec. III.C involve sorting and neighbor search, which are not strictly linear in the number of points.
  6. [Sec. V.D / Fig. 7] The real-world evaluation would be strengthened by a quantitative assessment of uniformity or detail preservation, since the current visual comparison is only qualitative.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported gains come from held-out benchmarks and the self-citations are background, not load-bearing.

full rationale

The paper's derivation chain is self-contained against external benchmarks. The single-photon point cloud is defined from the counting histogram (Eqs. (1)-(6)), the network is trained with the composite loss (Eq. (10)), and performance is measured on held-out PU-GAN and PU1K test splits with the same CD/HD/P2F metrics. While CD and P2F appear both in the loss and as evaluation metrics, the test sets are not used for training and the baselines are trained under the same protocol, so this is standard supervised evaluation rather than a prediction forced by construction. The Gaussian depth-offset corruption (Sec. IV) is a synthetic augmentation, not a parameter fitted to the target result, and the robustness claim is an empirical comparison on test data. Self-citations [23,26] support background statements about single-photon imaging and downstream tasks; they do not carry the upsampling or robustness claims. The paper itself acknowledges that the real-world comparison has no ground truth (Sec. V.D), which limits that evidence but is not circularity. The unsupported 'first to establish the upsampling framework' novelty claim and the lack of single-photon-specific baselines are external-validity/completeness weaknesses, not circularity.

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

No new physical entities are introduced. The novel components (multi-path scanning, bidirectional Mamba blocks, adaptive upsample shift) are software architectures that rearrange existing operators (edge conv, SSM, KPConv).

free parameters (5)
  • upsampling rate r_c = 4
    Set by hand and fixed for all experiments; the size of the output cloud depends linearly on it.
  • distortion noise level sigma (test) = 0, 0.02, 0.06, 0.1
    Hand-selected values for the synthetic Gaussian depth offset; the robustness claim is measured against these choices.
  • neighborhood size K in upsample shift
    Number of nearest neighbors used for deformable KPConv in Sec. III C; not specified numerically.
  • tunable radius R_k
    Radius for sampling kernel points from a Fibonacci sphere in Sec. III C; not specified.
  • loss distance scaling sigma
    Appears in the surface fitting loss (13) and repulsion loss (14); not specified as a hyperparameter value.
assumptions (5)
  • standard math Poisson photon-counting model h_ij ~ P(N(eta r_ij + B_d)) from Eq. (3) reproduces single-photon histogram statistics.
    Cited to [39]; the paper uses it to define histograms but the upsampling network does not depend on its exact form.
  • ad hoc to paper Adding zero-mean Gaussian depth offsets to generic CAD point clouds mimics real single-photon distortion.
    Introduced in Sec. IV to connect PU-GAN/PU1K to single-photon sensing; unvalidated against measured distortion.
  • domain assumption Mamba/state-space models with sequential scanning capture long-range dependencies in point clouds at linear cost.
    Taken from cited works [31-33]; no proof is offered in this paper.
  • domain assumption Chamfer distance, Hausdorff distance, and point-to-surface distance measure the reconstruction quality relevant to single-photon sensing.
    Standard in the PU literature; the paper does not justify their adequacy for photon-noise and depth-bias artifacts.
  • domain assumption Visual comparison on real single-photon LiDAR data is sufficient to establish detail preservation and noise suppression.
    Sec. V D presents only qualitative results because no ground truth exists; this assumes visual assessment is a reliable indicator.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Point upsampling networks for single-photon sensing." pith.science (2026). https://pith.science/paper/DP34RZAS

@misc{pith2026250812986,
  author       = {Pith},
  title        = {Pith review of: Point upsampling networks for single-photon sensing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DP34RZAS}},
  note         = {Machine review of arXiv:2508.12986}
}
read the original abstract

Single-photon sensing has generated great interest as a prominent technique of long-distance and ultra-sensitive imaging, however, it tends to yield sparse and spatially biased point clouds, thus limiting its practical utility. In this work, we propose using point upsampling networks to increase point density and reduce spatial distortion in single-photon point cloud. Particularly, our network is built on the state space model which integrates a multi-path scanning mechanism to enrich spatial context, a bidirectional Mamba backbone to capture global geometry and local details, and an adaptive upsample shift module to correct offset-induced distortions. Extensive experiments are implemented on commonly-used datasets to confirm its high reconstruction accuracy and strong robustness to the distortion noise, and also on real-world data to demonstrate that our model is able to generate visually consistent, detail-preserving, and noise suppressed point clouds. Our work is the first to establish the upsampling framework for single-photon sensing, and hence opens a new avenue for single-photon sensing and its practical applications in the downstreaming tasks.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

56 extracted references · 52 canonical work pages

  1. [1]

    Kirmani, D

    A. Kirmani, D. Venkatraman, D. Shin, A. Cola¸ co, F. N. Wong, J. H. Shapiro, and V. K. Goyal, First-photon imaging,Science343(6166), 58–61 (2014)

    work page 2014

  2. [40]

    J. Peng, Z. Xiong, X. Huang, Z.-P. Li, D. Liu, and F. Xu, Photon-efficient 3d imaging with a non-local neural network, inProc. Eur. Conf. Comput. Vis.2020, pp. 225– 241

    work page 2020

  3. [41]

    Z. Zang, D. Xiao, and D. D.-U. Li, Non-fusion time- resolved depth image reconstruction using a highly ef- ficient neural network architecture,Opt. Express29(13), 19278–19291 (2021)

    work page 2021

  4. [42]

    G. Yao, Y. Chen, Y. Liu, X. Hu, and Y. Pan, Ro- bust photon-efficient imaging using a pixel-wise resid- ual shrinkage network,Opt. Express30(11), 18856–18873 (2022)

    work page 2022

  5. [2]

    Defienne, W

    H. Defienne, W. P. Bowen, M. Chekhova, G. B. Lemos, D. Oron, S. Ramelow, N. Treps, and D. Faccio, Advances in quantum imaging,Nat. Photon. 18(10), 1024–1036 (2024)

    work page 2024

  6. [3]

    Y. Wang, K. Huang, J. Fang, M. Yan, E. Wu, and H. Zeng, Mid-infrared single-pixel imaging at the single- photon level,Nat. Commun.14(1), 1073 (2023)

    work page 2023

  7. [4]

    J. Lee, A. Ingle, J. V. Chacko, K. W. Eliceiri, and M. Gupta, CASPI: collaborative photon processing for ac- tive single-photon imaging,Nat. Commun.14(1), 3158 (2023)

    work page 2023

  8. [5]

    Vidyapin, Y

    V. Vidyapin, Y. Zhang, D. England, and B. Sussman, Characterisation of a single photon event camera for quantum imaging,Sci. Rep. 13(1), 1009 (2023)

    work page 2023

Show all 56 references
  1. [6]

    R. H. Hadfield, Single-photon detectors for optical quan- tum information applications,Nat. Photonics3(12), 696– 705 (2009)

    2009

  2. [7]

    Piron, D

    F. Piron, D. Morrison, M. R. Yuce, and J.-M. Redout’e, A review of single-photon avalanche diode time-of-flight imaging sensor arrays,IEEE Sens. J.21(11), 12654– 12666 (2020)

    2020

  3. [8]

    Y. Chen, G. Yao, Y. Liu, H. Su, X. Hu, and Y. Pan, Deep domain adversarial adaptation for photon-efficient imaging,Phys. Rev. Appl. 18(5), 054048 (2022)

    2022

  4. [9]

    D. Shin, A. Kirmani, V. K. Goyal, and J. H. Shapiro, Photon-efficient computational 3-D and reflectivity imag- ing with single-photon detectors,IEEE Trans. Comput. Imaging1(2), 112–125 (2015)

    2015

  5. [10]

    Bruschini, H

    C. Bruschini, H. Homulle, I. M. Antolovic, S. Burri, and E. Charbon, Single-photon avalanche diode imagers in biophotonics: review and outlook,Light Sci. Appl.8(1), 87 (2019)

    2019

  6. [11]

    Bocchieri, E

    A. Bocchieri, E. Charbon, and A. Velten, Scintillation event imaging with a single photon avalanche diode cam- era,Commun. Eng.3(1), 135 (2024)

    2024

  7. [12]

    P. J. De Visser, S. A. De Rooij, V. Murugesan, D. J. Thoen, and J. J. Baselmans, Phonon-trapping-enhanced energy resolution in superconducting single-photon de- tectors,Phys. Rev. Appl.16(3), 034051 (2021)

    2021

  8. [13]

    Maccarone, A

    A. Maccarone, A. McCarthy, X. Ren, R. E. Warbur- ton, A. M. Wallace, J. Moffat, Y. Petillot, and G. S. Buller, Underwater depth imaging using time-correlated single-photon counting,Opt. Express23(26), 33911– 33926 (2015)

    2015

  9. [14]

    H. Shi, G. Shen, H. Qi, Q. Zhan, H. Pan, Z. Li, and G. Wu, Noise-tolerant Bessel-beam single-photon imaging in fog,Opt. Express30(7), 12061–12068 (2022)

    2022

  10. [15]

    Zhang, S

    Y. Zhang, S. Li, J. Sun, X. Zhang, D. Liu, X. Zhou, H. Li, and Y. Hou, Three-dimensional single-photon imaging through realistic fog in an outdoor environment during the day,Opt. Express30(19), 34497–34509 (2022)

    2022

  11. [16]

    A. M. Pawlikowska, A. Halimi, R. A. Lamb, and G. S. Buller, Single-photon three-dimensional imaging at up to 10 kilometers range,Opt. Express25(10), 11919–11931 (2017)

    2017

  12. [17]

    Z.-P. Li, X. Huang, P.-Y. Jiang, Y. Hong, C. Yu, Y. Cao, J. Zhang, F. Xu, and J.-W. Pan, Super-resolution single- photon imaging at 8.2 kilometers,Opt. Express28(3), 4076–4087 (2020)

    2020

  13. [18]

    Z.-P. Li, X. Huang, Y. Cao, B. Wang, Y.-H. Li, W. Jin, C. Yu, J. Zhang, Q. Zhang, C.-Z. Peng et al., Single-photon computational 3D imaging at 45 km,Photon. Res.8(9), 1532–1540 (2020)

    2020

  14. [19]

    Li, J.-T

    Z.-P. Li, J.-T. Ye, X. Huang, P.-Y. Jiang, Y. Cao, Y. Hong, C. Yu, J. Zhang, Q. Zhang, C.-Z. Peng et al., Single-photon imaging over 200 km,Optica8(3), 344– 349 (2021)

    2021

  15. [20]

    R. H. Hadfield, J. Leach, F. Fleming, D. J. Paul, C. H. Tan, J. S. Ng, R. K. Henderson, and G. S. Buller, Single- photon detection for long-range imaging and sensing,Op- tica10(9), 1124–1141 (2023)

    2023

  16. [21]

    Madonini, F

    F. Madonini, F. Severini, F. Zappa, and F. Villa, Sin- gle photon avalanche diode arrays for quantum imaging and microscopy,Adv. Quantum Technol.4(7), 2100005 (2021)

    2021

  17. [22]

    Maccarone, G

    A. Maccarone, G. Acconcia, U. Steinlehner, I. Labanca, D. Newborough, I. Rech, and G. S. Buller, Custom- 11 technology single-photon avalanche diode linear detec- tor array for underwater depth imaging,Sensors21(14), 4850 (2021)

    2021

  18. [23]

    Y. Hong, S. Liu, Z.-P. Li, X. Huang, P. Jiang, Y. Xu, C. Wu, H. Zhou, Y.-C. Zhang, H.-L. Ren et al., Airborne single-photon lidar towards a small-sized and low-power payload,Optica11(5), 612–618 (2024)

    2024

  19. [24]

    Rapp and V

    J. Rapp and V. K. Goyal, A few photons among many: Unmixing signal and noise for photon-efficient active imaging,IEEE Trans. Comput. Imaging3(3), 445–459 (2017)

    2017

  20. [25]

    Ma, Y.-S

    B. Ma, Y.-S. Liu, and Z. Han, Reconstructing surfaces for sparse point clouds with on-surface priors, inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2022, pp. 6315–6325

    2022

  21. [26]

    X. Li, J. Liu, G. Zhao, L. Liu, W. Zhang, X. Hu, and S. Cheng, High precision single-photon object detection via deep neural networks,Opt. Express, 32(21), 37224–37237 (2024)

    2024

  22. [27]

    Zheng, Y

    Y. Zheng, Y. Li, S. Yang, and H. Lu, Global-pbnet: A novel point cloud registration for autonomous driving, IEEE Trans. Intell. Transp. Syst.23(12), 22312–22319 (2022)

    2022

  23. [28]

    Li and G

    J. Li and G. H. Lee, Deepi2p: Image-to-point cloud regis- tration via deep classification, inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2021, pp. 15960–15969

    2021

  24. [29]

    S. Qiu, S. Anwar, and N. Barnes, Dense-resolution net- work for point cloud classification and segmentation, inProc. IEEE/CVF Winter Conf. Appl. Comput. Vis., 2021, pp. 3813–3822

    2021

  25. [30]

    S. Qiu, Y. Wu, S. Anwar, and C. Li, Investigating at- tention mechanism in 3d point cloud object detection, in Proc. Int. Conf. 3D Vis., 2021, pp. 403–412

    2021

  26. [31]

    Gu and T

    A. Gu and T. Dao, Mamba: Linear-time sequence modeling with selective state spaces,arXiv preprint arXiv:2312.00752(2023)

    2023 arXiv

  27. [32]

    Zhang, X

    T. Zhang, X. Li, H. Yuan, S. Ji, and S. Yan, Point cloud mamba: Point cloud learning via state space model, arXiv preprint arXiv:2403.00762(2024)

    2024 arXiv

  28. [34]

    L. Yu, X. Li, C.-W. Fu, D. Cohen-Or, and P.-A. Heng, Pu-net: Point cloud upsampling network, inProc. IEEE Conf. Comput. Vis. Pattern Recognit., 2018, pp. 2790– 2799

    2018

  29. [35]

    G. Qian, A. Abualshour, G. Li, A. Thabet, and B. Ghanem, Pu-gcn: Point cloud upsampling using graph convolutional networks, inProc. IEEE/CVF Conf. Com- put. Vis. Pattern Recognit., 2021, pp. 11683–11692

    2021

  30. [36]

    Yifan, S

    W. Yifan, S. Wu, H. Huang, D. Cohen-Or, and O. Sorkine-Hornung, Patch-based progressive 3d point set upsampling, inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2019, pp. 5958–5967

    2019

  31. [37]

    Y. He, D. Tang, Y. Zhang, X. Xue, and Y. Fu, Grad- pu: Arbitrary-scale point cloud upsampling via gradi- ent descent with learned distance functions, inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2023, pp. 5354–5363

    2023

  32. [38]

    S. Qiu, S. Anwar, and N. Barnes, Pu-transformer: Point cloud upsampling transformer, inProc. Asian Conf. Comput. Vis., 2022, pp. 2475–2493

    2022

  33. [39]

    D. Shin, J. H. Shapiro, and V. K. Goyal, Photon-efficient super-resolution laser radar, inProc. Eur. Conf. Comput. Vis., vol. 10394, 2017, pp. 9–16

    2017

  34. [43]

    Scholes, G

    S. Scholes, G. Mora-Martin, F. Zhu, I. Gyongy, P. Soan, and J. Leach, Fundamental limits to depth imaging with single-photon detector array sensors,Sci. Rep.13(1), 176 (2023)

    2023

  35. [44]

    Vaswani, Attention is all you need,Adv

    A. Vaswani, Attention is all you need,Adv. Neural Inf. Process. Syst.30, (2017)

    2017

  36. [45]

    Y. Wang, Y. Sun, Z. Liu, S. E. Sarma, M. M. Bronstein, and J. M. Solomon, Dynamic graph CNN for learning on point clouds,ACM Trans. Graphics38(5), 1–12 (2019)

    2019

  37. [46]

    J. Liu, R. Yu, Y. Wang, Y. Zheng, T. Deng, W. Ye, and H. Wang, Point mamba: A novel point cloud backbone based on state space model with octree-based ordering strategy,arXiv preprint arXiv:2403.06467(2024)

    2024 arXiv

  38. [47]

    Thomas, C

    H. Thomas, C. R. Qi, J.-E. Deschaud, B. Marcotegui, F. Goulette, and L. J. Guibas, Kpconv: Flexible and deformable convolution for point clouds, inProc. IEEE Int. Conf. Comput. Vis., 2019, pp. 6411-6420

    2019

  39. [48]

    R. Li, X. Li, C.-W. Fu, D. Cohen-Or, P.-A. Heng, Pu-gan: a point cloud upsampling adversarial network, inProc. IEEE Int. Conf. Comput. Vis., 2019, pp. 7203-7212

    2019

  40. [49]

    A. X. Chang, T. Funkhouser, L. Guibas, P. Hanrahan, Q. Huang, Z. Li, S. Savarese, M. Savva, S. Song, H. Su et al., Shapenet: An information-rich 3d model repository, arXiv preprint arXiv:1512.03012(2015)

    2015 arXiv

  41. [50]

    R. Li, X. Li, P.-A. Heng, and C.-W. Fu, Point cloud upsampling via disentangled refinement, inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2021, pp. 344-353

    2021

  42. [51]

    Y. Rong, H. Zhou, K. Xia, C. Mei, J. Wang, and T. Lu, Repkpu: Point cloud upsampling with kernel point rep- resentation and deformation, inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2024, pp. 21050-21060

    2024

  43. [52]

    Kim and Y

    K. Kim and Y. Lee, Improvement of signal and noise performance using single image super-resolution based on deep learning in single photon-emission computed to- mography imaging system,Nucl. Eng. Technol.53(7), 2341-2347 (2021)

    2021

  44. [53]

    Laurenzis, T

    M. Laurenzis, T. Seets, E. Bacher, A. Ingle, and A. Vel- ten, Comparison of super-resolution and noise reduction for passive single-photon imaging,J. Electron. Imaging 31(3), 033042 (2022)

    2022

  45. [54]

    Y. Liu, F. Gutierrez-Barragan, A. Ingle, M. Gupta, and A. Velten, Single-photon camera guided extreme dy- namic range imaging, inProc. IEEE/CVF Winter Conf. Appl. Comput. Vis., 2022, pp. 1575-1585

    2022

  46. [55]

    L. Bian, H. Song, L. Peng, X. Chang, X. Yang, R. Horstmeyer, L. Ye, C. Zhu, T. Qin, D. Zheng et al., High- resolution single-photon imaging with physics-informed deep learning,Nat. Commun.14(1), 5902 (2023). 13 state equation and the observation equation. Formally, an SSM is g...

    2023

  47. [56]

    Discretization for Point Sequence Processing Discrete features along scan paths are processed by applying a zero-order hold (ZOH) discretization method [31] with a time step∆. The discretization trans- forms the continuous parametersAandBinto discrete counterparts ¯Aand ¯B: ht...

  48. [57]

    This mechanism adapts to the local context of point clouds by modulating transitions and readouts based on input features

    Mamba’s Selective Mechanism for Point Clouds Mamba introduces a selective scanning mechanism that enhances traditional SSM [32, 33]. This mechanism adapts to the local context of point clouds by modulating transitions and readouts based on input features. The mechanism efficie...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.