Pith. sign in

REVIEW 3 major objections 6 minor 49 references

Towards Generalized Range-View LiDAR Segmentation in Adverse Weather

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Two stem modules lift range-view LiDAR segmentation from 7.9 to 28.2 mIoU under adverse weather.

desk verdict Solid plug-in for range-view LiDAR under weather shift; the GAS transfer assumption is the one real soft spot, but the method stands on reflectance separation/calibration. read the letter →

arxiv 2506.08979 v3 pith:OHA2C3V4 submitted 2025-06-10 cs.CV cs.RO

Longyu Yang , Lu Zhang , Jun Liu , Yap-Peng Tan , Heng Tao Shen , Xiaofeng Zhu , Ping Hu This is my paper
classification cs.CVcs.RO
keywords LiDARsemanticsegmentationrange-viewrepresentationadverseweatherdomaingeneralizationgeometricabnormalitysuppressionreflectancedistortioncalibrationmemory-guidednormalizationautonomousdriving
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 claims that range-view LiDAR semantic segmentation models lose much of their accuracy in rain, snow, and fog because the range-view projection concentrates weather-induced noise and reflectance distortion into dense 2D pixels, and that a stem-level intervention can fix this. Its proposed framework splits the network's initial stem into two branches: a Geometric Abnormality Suppression (GAS) module trained self-supervised on clean geometry to down-weight noise-like features, and a Reflectance Distortion Calibration (RDC) module that re-aligns reflectance statistics to clean-weather style through a memory bank. On SemanticKITTI→SemanticSTF, adding the two modules to SalsaNext raises mIoU from 7.9 to 28.2 and to RangeViT from 10.2 to 28.9, while source-domain accuracy is preserved and inference latency grows by about 3 ms. If correct, this makes robust adverse-weather segmentation achievable as a lightweight, architecture-agnostic add-on rather than a full redesign.

What carries the argument

The central machinery is a dual-branch stem replacement: range-view inputs are split into geometric attributes and reflectance intensity, processed separately, then fused and fed to the original backbone. GAS is a lightweight abnormality classifier whose positives are clean geometric features and whose negatives are standard Gaussian feature maps, trained with cross-entropy on $F^+ = F_{\text{geo}} + \gamma\epsilon$ ($\gamma=0.02$) versus $f^- \sim \mathcal{N}(0,I)$; at inference the softmax output $W^+$ weights the features as $\hat{F}_{\text{geo}} = W^+ \odot F_{\text{geo}}$. RDC performs memory-guided adaptive instance normalization: a Source Style Memory Bank $\mathbf{M}\in\mathbb{R}^{T\times C}$ is queried by cosine similarity, a softmax over $T$ retrieves per-pixel style vectors $V = \mathbf{M}^T A$, the retrieved channel-wise mean/variance serve as source statistics, and the reflectance feature is re-centered and re-scaled: $\hat{F}_{\text{ref}} = \sigma_{\text{src}}(F_{\text{ref}}) \, (F_{\text{ref}}-\mu_{\text{ref}})/\sigma_{\text{ref}} + \mu_{\text{src}}(F_{\text{ref}})$, with random perturbation of input statistics during training for robustness. The two branches are fused by point-wise summation, and the total loss is $L = L_{\text{GAS}} + L_{\text{RDC}} + L_{\text{SEG}}$.

What would settle it

On SemanticSTF validation scans, compare the GAS classifier's normality map $W^+$ against hand-annotated masks of weather-induced spurious returns (e.g., raindrop backscatter and snowflake points) and of valid distant surfaces. If $W^+$ does not preferentially down-weight the spurious-return pixels while leaving valid geometry intact, the self-supervised Gaussian decision boundary is not transferring to real weather artifacts, and the reported mIoU gains must be attributed to the dual-branch stem or RDC rather than to geometric abnormality suppression.

Watch

Extended reading notes

Core claim

The central claim is that weather-induced failure in range-view LiDAR segmentation decomposes into two correctable phenomena: geometric noise from spurious returns (raindrop backscatter, snowflakes, fog Mie scattering) and systematic distortion of reflectance intensity, both amplified by the 2D projection. GAS addresses the first by learning, entirely on clean source data, a binary classifier that separates structured geometric features from isotropic Gaussian noise; at inference its softmax 'normality' map multiplies the geometric features so anomalous patterns are suppressed. RDC addresses the second by storing clean-weather style vectors in a learnable memory bank and, per pixel, retrieving the most similar source styles via cosine similarity, then normalizing the reflectance feature map to match those retrieved channel-wise statistics. Trained with a semantic-consistency loss and a style-alignment loss plus standard source-domain augmentations, the two modules are inserted only into the stem block, leaving the backbone and segmentation head untouched, and are demonstrated across four range-view backbones and three transfer settings.

Load-bearing premise

The load-bearing premise is that a binary classifier trained only on the clean source domain to separate structured geometric features from pure Gaussian noise will, at test time, recognize real rain-, snow-, and fog-induced geometric artifacts as abnormal and down-weight them correctly.

Editorial extensions

If this is right

  • Range-view models equipped with GAS and RDC reach or exceed voxel-based generalization: RangeViT reaches 28.9 mIoU on SemanticSTF, above the 28.6 reported for the voxel-based MinkNet+PDR baseline.
  • The remedy is architecture-agnostic: the same stem replacement improves SalsaNext, RangeNet++, CENet, and RangeViT, and a preliminary MinkNet experiment indicates the modules also transfer to voxel backbones.
  • Training requires only clean-weather source data and standard augmentations; no weather simulation, paired adverse-weather data, or target-domain labels are needed.
  • Inference overhead is small enough for latency-sensitive use: roughly 3 ms added latency and 0.1 M added parameters on the evaluated backbones, with no loss of source-domain accuracy.

Reading between the lines

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

  • The GAS decision boundary was learned against isotropic Gaussian negatives; real weather artifacts are range-dependent and often anisotropic, so the key transferable assumption is that 'abnormal' geometry in the source domain is a good proxy for weather-corrupted geometry, which the ablation quantifies but does not isolate.
  • The paper compares against PointDR and RDA but does not test whether GAS and RDC are complementary to augmentation-based methods; combining stem-level calibration with source-domain augmentation is a natural next experiment.
  • Since the modules operate on raw range-view inputs before any backbone-specific feature extraction, a direct extension is to apply the same stem swap to range-view object detection or panoptic segmentation models, where adverse-weather robustness is equally critical.
  • The resolution study suggests that in adverse weather the baseline degrades with higher resolution while the proposed method improves; this hints that the stem swap may change the effective noise-concentration trade-offs and could be combined with resolution-adaptive inference.
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 / 6 minor

Summary. The paper proposes a modular framework to improve the generalization of range-view LiDAR semantic segmentation models to adverse weather. The initial stem block of a range-view network is split into two branches: a Geometric Abnormality Suppression (GAS) module, which uses a binary classifier trained on clean geometric features and synthetic Gaussian noise to down-weight weather-induced geometric anomalies, and a Reflectance Distortion Calibration (RDC) module, which uses memory-guided adaptive instance normalization to correct reflectance distortions toward source-domain statistics. The method is applied to four range-view baselines (SalsaNext, RangeNet++, CENet, RangeViT) trained on SemanticKITTI or SynLiDAR and evaluated on SemanticSTF and SemanticKITTI-C, reporting large mIoU gains (e.g., SalsaNext 7.9 to 28.2 on SemanticSTF) with modest latency increases.

Significance. If the proposed mechanisms work as intended, the paper makes a practical contribution: it improves adverse-weather generalization without requiring weather simulation or target-domain data, is modular across architectures, and preserves source-domain accuracy. The evaluation is extensive, covering four baselines, two target datasets, ablations, and efficiency measurements, and the design is clearly motivated by the specific difficulties of range-view representations. However, the central mechanism of GAS relies on an untested transfer assumption, and the RDC losses have a potential trivial-solution problem. These issues are load-bearing for the claimed improvements, so the current evidence is conditional on additional validation.

major comments (3)
  1. [Section 3.2, Eqs. (2)-(3) and (7), Table 4] The GAS abnormality classifier is trained to distinguish clean geometric features from i.i.d. Gaussian noise, but at inference it is used to gate features corrupted by real rain, snow, and fog. Weather artifacts are spatially structured, often concentrated in specific beams or regions, and mixed with valid structure in the same pixel, whereas the training negatives are white noise. The paper provides no evidence that the learned decision boundary transfers to these real artifacts: there is no analysis of W^+ on SemanticSTF, no ablation replacing W^+ with a constant gate, and no sensitivity study of the noise scale gamma. Given that GAS contributes a substantial share of the reported gain (Table 4: +3.9 mIoU from 20.1 to 24.0 over the geometry-only stem, and +12.4 mIoU from 15.8 to 28.2 when combined with RDC), this untested transfer is load-bearing for the central generalization claim. The authors should add experiments that directly evaluate the behavior of W^+ on target weather data and compare against a constant-gating baseline.
  2. [Section 3.3, Eqs. (13)-(14)] The semantic consistency loss LSC minimizes the difference between the stylized augmented feature and the original, and the style alignment loss LSA minimizes the difference between the retrieved source statistics and the input statistics. Minimizing both encourages the memory bank to return the input's own statistics, which is the trivial solution and would make RDC an identity mapping that cannot correct weather-induced reflectance drift. Although the segmentation loss in Eq. (16) may prevent complete collapse, the paper does not demonstrate that the memory bank learns non-trivial transferable styles. Please provide an analysis of the retrieved style statistics on source versus target data, and an ablation that removes LSC and LSA to quantify their contribution to the final performance.
  3. [Section 4, Tables 1-2] All experimental results are reported from a single run with no standard deviations or confidence intervals. This is particularly problematic for the SemanticKITTI-C benchmark, where improvements are modest (e.g., +1.6 mIoU for CENet), and for the source-domain accuracy claims, where some models show slight decreases (e.g., CENet from 60.9 to 60.5, RangeViT from 59.6 to 59.3). The paper's assertions that source-domain accuracy is preserved and that the method consistently improves generalization rest on these numbers; without error bars the reader cannot assess statistical reliability. Please report results over multiple random seeds.
minor comments (6)
  1. [Eq. (6)] The subscripts M in S^+_M and S^-_M are not defined; please clarify the notation.
  2. [Eqs. (2)-(3)] The notation for the positive and negative samples is inconsistent: F^+ in Eq. (2) but f^- in Eq. (3); please unify.
  3. [Table 1] The table headers appear garbled in the compiled version (long category names run together), and 'RangeNet ++' contains an extra space; please reformat.
  4. [Section 4.1] In the Implementation Details, 'repalce' should be 'replace'.
  5. [Section 4.3, Table 3] The claimed 'minimal inference overhead' is not normalized by baseline latency; for SalsaNext the increase is 2.9 ms (29%), for CENet 4.7 ms (25%), and for RangeNet++ 0.7 ms (4%). Reporting the relative overhead would strengthen the efficiency claim.
  6. [Section 4.2] The comparison with GRC [47] is described without noting that it is a concurrent work from the same group; please add a statement for transparency.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular steps: GAS and RDC are trained on the source domain and evaluated on unseen adverse-weather benchmarks; the only self-citation (GRC [47]) is a non-load-bearing comparison baseline.

full rationale

The derivation chain is not circular. GAS is trained on a synthetic pretext task defined entirely from source-domain geometry (Eqs. 2-6) and at inference multiplies the geometric feature map by the classifier's normal-probability map (Eq. 7); RDC is trained on source-domain reflectance only, using a memory bank and augmentation losses (Eqs. 8-14). Neither module uses SemanticSTF or SemanticKITTI-C during training, and the reported improvements (e.g., SalsaNext 7.9 to 28.2 mIoU on SemanticKITTI->SemanticSTF, Table 1) are measured on these unseen target benchmarks. The GAS classifier's transfer from Gaussian negatives to real rain, snow, and fog is an empirical generalization assumption, not a circular reduction: it is not fitted to the target, and no target statistic is renamed as a prediction. The RDC losses do encourage the calibrated features to remain close to source inputs (Eqs. 13-14), but that is a source-domain regularization strategy rather than a target-domain fit. The only self-citation, GRC [47], is used as a comparison baseline adapted to the range-view setting, not as a load-bearing premise or uniqueness theorem. No load-bearing step reduces by construction to its own input; the score reflects a minor non-load-bearing self-citation amid otherwise self-contained evaluation.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The proposed approach depends on three premises: (1) GAS's self-supervised boundary against Gaussian noise transfers to real weather noise; (2) aligning reflectance statistics to the source domain is the correct fix for weather-induced distortion; (3) the two-branch design with simple summation fusion preserves useful information. The method introduces few hand-set free parameters; the main unvalidated load is premise (1).

free parameters (2)
  • gamma = 0.02
    Perturbation scale for positive samples in GAS (Eq. 2); chosen empirically.
  • alpha, beta = sampled from U(0,1) per channel
    Random scales for channel statistics augmentation in RDC (Eq. 11); stochastic training hyperparameters.
assumptions (3)
  • domain assumption Manifold hypothesis for geometric features
    GAS assumes clean geometric features lie on a low-dimensional manifold and weather noise deviates from it (Section 3.2, citing Refs. 44, 45).
  • ad hoc to paper Isotropic Gaussian noise is a valid surrogate for weather-induced geometric noise in feature space
    Negative samples in GAS training are standard Gaussian (Eq. 3); this assumes the decision boundary learned against Gaussian noise transfers to real weather artifacts.
  • domain assumption Aligning reflectance statistics to source-domain style is the correct reversal of weather-induced distortion
    RDC's memory-guided AdaIN (Section 3.3) assumes source statistics are the desired target for calibration under all weather conditions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Generalized Range-View LiDAR Segmentation in Adverse Weather." pith.science (2026). https://pith.science/paper/OHA2C3V4

@misc{pith2026250608979,
  author       = {Pith},
  title        = {Pith review of: Towards Generalized Range-View LiDAR Segmentation in Adverse Weather},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OHA2C3V4}},
  note         = {Machine review of arXiv:2506.08979}
}
read the original abstract

LiDAR segmentation has emerged as an important task to enrich scene perception and understanding. Range-view-based methods have gained popularity due to their high computational efficiency and compatibility with real-time deployment. However, their generalized performance under adverse weather conditions remains underexplored, limiting their reliability in real-world environments. In this work, we identify and analyze the unique challenges that affect the generalization of range-view LiDAR segmentation in severe weather. To address these challenges, we propose a modular and lightweight framework that enhances robustness without altering the core architecture of existing models. Our method reformulates the initial stem block of standard range-view networks into two branches to process geometric attributes and reflectance intensity separately. Specifically, a Geometric Abnormality Suppression (GAS) module reduces the influence of weather-induced spatial noise, and a Reflectance Distortion Calibration (RDC) module corrects reflectance distortions through memory-guided adaptive instance normalization. The processed features are then fused and passed to the original segmentation pipeline. Extensive experiments on different benchmarks and baseline models demonstrate that our approach significantly improves generalization to adverse weather with minimal inference overhead, offering a practical and effective solution for real-world LiDAR segmentation.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

49 extracted references · 40 canonical work pages

  1. [1]

    In: CVPR (2017)

    Qi, C.R., Su, H., Mo, K., Guibas, L.J.: Pointnet: Deep learning on point sets for 3d classification and segmentation. In: CVPR (2017)

    2017

  2. [2]

    IEEE Trans

    Hu, Q., Yang, B., Xie, L., Rosa, S., Guo, Y., Wang, Z., Trigoni, N., Markham, A.: Learning semantic segmentation of large- scale point clouds with random sampling. IEEE Trans. Pattern Anal. Mach. Intell. 44(11), 8338–8354 (2021)

    work page 2021

  3. [3]

    In: ICCV (2023)

    Kong, L., Liu, Y., Chen, R., Ma, Y., Zhu, X., Li, Y., Hou, Y., Qiao, Y., Liu, Z.: Rethinking range view representation for lidar segmenta- tion. In: ICCV (2023)

    work page 2023

  4. [4]

    Wu, X., Lao, Y., Jiang, L., Liu, X., Zhao, H.: Point transformer v2: Grouped vector atten- tion and partition-based pooling. Adv. Neu- ral Inform. Process. Syst. 35, 33330–33342 (2022)

    work page 2022

  5. [5]

    In: CVPR (2024)

    Wu, X., Jiang, L., Wang, P.-S., Liu, Z., Liu, X., Qiao, Y., Ouyang, W., He, T., Zhao, H.: Point transformer v3: Simpler faster stronger. In: CVPR (2024)

    work page 2024

  6. [6]

    In: CVPR (2019)

    Choy, C., Gwak, J., Savarese, S.: 4d spatio- temporal convnets: Minkowski convolutional neural networks. In: CVPR (2019)

    work page 2019

  7. [7]

    In: ICCV (2021)

    Zhou, H., Zhu, X., Song, X., Ma, Y., Wang, Z., Li, H., Lin, D.: Cylinder3d: An effective 3d framework for driving-scene lidar semantic segmentation. In: ICCV (2021)

    work page 2021

  8. [8]

    arXiv preprint arXiv:2304.06906 (2023)

    Yang, Y.-Q., Guo, Y.-X., Xiong, J.-Y., Liu, Y., Pan, H., Wang, P.-S., Tong, X., Guo, B.: Swin3d: A pretrained transformer backbone for 3d indoor scene understanding. arXiv preprint arXiv:2304.06906 (2023)

    arXiv 2023

Show all 49 references
  1. [9]

    In: ECCV (2022)

    Choe, J., Park, C., Rameau, F., Park, J., Kweon, I.S.: Pointmixer: Mlp-mixer for point cloud understanding. In: ECCV (2022)

    2022

  2. [10]

    arXiv preprint arXiv:2410.04960 (2024)

    Sun, X., Liu, J., Shen, H.T., Zhu, X., Hu, P.: On efficient variants of segment anything model: A survey. arXiv preprint arXiv:2410.04960 (2024)

    2024 arXiv

  3. [11]

    In: ICCV (2019) 13

    Behley, J., Garbade, M., Milioto, A., Quen- zel, J., Behnke, S., Stachniss, C., Gall, J.: Semantickitti: A dataset for semantic scene understanding of lidar sequences. In: ICCV (2019) 13

    2019

  4. [12]

    In: CVPR (2020)

    Caesar, H., Bankiti, V., Lang, A.H., Vora, S., Liong, V.E., Xu, Q., Krishnan, A., Pan, Y., Baldan, G., Beijbom, O.: nuscenes: A mul- timodal dataset for autonomous driving. In: CVPR (2020)

    2020

  5. [13]

    In: ICCV (2021)

    Hahner, M., Sakaridis, C., Dai, D., Van Gool, L.: Fog simulation on real lidar point clouds for 3d object detection in adverse weather. In: ICCV (2021)

    2021

  6. [14]

    In: CVPR (2020)

    Bijelic, M., Gruber, T., Mannan, F., Kraus, F., Ritter, W., Dietmayer, K., Heide, F.: See- ing through fog without seeing fog: Deep multimodal sensor fusion in unseen adverse weather. In: CVPR (2020)

    2020

  7. [15]

    In: CVPR (2023)

    Xiao, A., Huang, J., Xuan, W., Ren, R., Liu, K., Guan, D., El Saddik, A., Lu, S., Xing, E.P.: 3d semantic segmentation in the wild: Learning generalized models for adverse-condition point clouds. In: CVPR (2023)

    2023

  8. [16]

    In: CVPR (2022)

    Hahner, M., Sakaridis, C., Bijelic, M., Heide, F., Yu, F., Dai, D., Van Gool, L.: Lidar snow- fall simulation for robust 3d object detection. In: CVPR (2022)

    2022

  9. [17]

    In: CVPR (2024)

    Zhao, H., Zhang, J., Chen, Z., Zhao, S., Tao, D.: Unimix: Towards domain adaptive and generalizable lidar semantic segmentation in adverse weather. In: CVPR (2024)

    2024

  10. [18]

    arXiv preprint arXiv:2312.12772 (2023)

    Yang, D., Liu, Z., Jiang, W., Yan, G., Gao, X., Shi, B., Liu, S., Cai, X.: Realistic rainy weather simulation for lidars in carla simula- tor. arXiv preprint arXiv:2312.12772 (2023)

    2023 arXiv

  11. [19]

    arXiv preprint arXiv:2407.02286 (2024)

    Park, J., Kim, K., Shim, H.: Rethinking data augmentation for robust lidar semantic seg- mentation in adverse weather. arXiv preprint arXiv:2407.02286 (2024)

    2024 arXiv

  12. [20]

    In: ACMMM (2024)

    He, P., Jiao, L., Li, L., Liu, X., Liu, F., Ma, W., Yang, S., Shang, R.: Domain generalization-aware uncertainty introspec- tive learning for 3d point clouds segmenta- tion. In: ACMMM (2024)

    2024

  13. [21]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference, pp

    Park, J., Lee, H., Kang, I., Shim, H.: No thing, nothing: Highlighting safety-critical classes for robust lidar semantic segmentation in adverse weather. In: Proceedings of the Computer Vision and Pattern Recognition Conference, pp. 6690–6699 (2025)

    2025

  14. [22]

    Qi, C.R., Yi, L., Su, H., Guibas, L.J.: Point- net++: Deep hierarchical feature learning on point sets in a metric space. Adv. Neural Inform. Process. Syst. 30 (2017)

    2017

  15. [23]

    In: ICCV (2019)

    Thomas, H., Qi, C.R., Deschaud, J.-E., Marcotegui, B., Goulette, F., Guibas, L.J.: Kpconv: Flexible and deformable convolution for point clouds. In: ICCV (2019)

    2019

  16. [24]

    In: CVPR (2019)

    Wu, W., Qi, Z., Fuxin, L.: Pointconv: Deep convolutional networks on 3d point clouds. In: CVPR (2019)

    2019

  17. [25]

    In: Computer Vision–ECCV 2020: 16th European Confer- ence, Glasgow, UK, August 23–28, 2020, Pro- ceedings, Part XXIII 16, pp

    Liu, Z., Hu, H., Cao, Y., Zhang, Z., Tong, X.: A closer look at local aggregation oper- ators in point cloud analysis. In: Computer Vision–ECCV 2020: 16th European Confer- ence, Glasgow, UK, August 23–28, 2020, Pro- ceedings, Part XXIII 16, pp. 326–342 (2020). Springer

    2020

  18. [26]

    ACM Trans

    Wang, Y., Sun, Y., Liu, Z., Sarma, S.E., Bronstein, M.M., Solomon, J.M.: Dynamic graph cnn for learning on point clouds. ACM Trans. Graph. 38(5), 1–12 (2019)

    2019

  19. [27]

    In: CVPR (2018)

    Landrieu, L., Simonovsky, M.: Large-scale point cloud semantic segmentation with superpoint graphs. In: CVPR (2018)

    2018

  20. [28]

    IEEE Transactions on Neural Networks and Learning Systems 35(4), 4798–4812 (2022)

    Du, Z., Ye, H., Cao, F.: A novel local–global graph convolutional method for point cloud semantic segmentation. IEEE Transactions on Neural Networks and Learning Systems 35(4), 4798–4812 (2022)

    2022

  21. [29]

    In: ICCV (2021)

    Zhao, H., Jiang, L., Jia, J., Torr, P.H., Koltun, V.: Point transformer. In: ICCV (2021)

    2021

  22. [30]

    In: CVPR (2022) 14

    Lai, X., Liu, J., Jiang, L., Wang, L., Zhao, H., Liu, S., Qi, X., Jia, J.: Stratified transformer for 3d point cloud segmentation. In: CVPR (2022) 14

    2022

  23. [31]

    Robert, D., Raguet, H., Landrieu, L.: Effi- cient 3d semantic segmentation with super- point transformer. arxiv. arXiv preprint arXiv:2306.08045 (2023)

    2023 arXiv

  24. [32]

    In: CVPR (2015)

    Wu, Z., Song, S., Khosla, A., Yu, F., Zhang, L., Tang, X., Xiao, J.: 3d shapenets: A deep representation for volumetric shapes. In: CVPR (2015)

    2015

  25. [33]

    In: CVPR (2023)

    Lai, X., Chen, Y., Lu, F., Liu, J., Jia, J.: Spherical transformer for lidar-based 3d recognition. In: CVPR (2023)

    2023

  26. [34]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, pp

    Han, L., Zheng, T., Xu, L., Fang, L.: Occuseg: Occupancy-aware 3d instance segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, pp. 2940–2949 (2020)

    2020

  27. [35]

    In: CVPR (2018)

    Graham, B., Engelcke, M., Van Der Maaten, L.: 3d semantic segmentation with submani- fold sparse convolutional networks. In: CVPR (2018)

    2018

  28. [36]

    In: ISVC (2020)

    Cortinhal, T., Tzelepis, G., Erdal Aksoy, E.: Salsanext: Fast, uncertainty-aware semantic segmentation of lidar point clouds. In: ISVC (2020)

    2020

  29. [37]

    In: IROS (2019)

    Milioto, A., Vizzo, I., Behley, J., Stachniss, C.: Rangenet++: Fast and accurate lidar semantic segmentation. In: IROS (2019)

    2019

  30. [38]

    In: ECCV (2025)

    Li, L., Shum, H.P., Breckon, T.P.: Rapid-seg: Range-aware pointwise distance distribution networks for 3d lidar segmentation. In: ECCV (2025)

    2025

  31. [39]

    In: CVPR (2023)

    Ando, A., Gidaris, S., Bursuc, A., Puy, G., Boulch, A., Marlet, R.: Rangevit: Towards vision transformers for 3d semantic segmen- tation in autonomous driving. In: CVPR (2023)

    2023

  32. [40]

    In: 2022 IEEE International Conference on Multimedia and Expo (ICME), pp

    Cheng, H.-X., Han, X.-F., Xiao, G.-Q.: Cenet: Toward concise and efficient lidar semantic segmentation for autonomous driv- ing. In: 2022 IEEE International Conference on Multimedia and Expo (ICME), pp. 01–06 (2022). IEEE

    2022

  33. [41]

    Yan, X., Zheng, C., Xue, Y., Li, Z., Cui, S., Dai, D.: Benchmarking the robustness of lidar semantic segmentation models. Int. J. Comput. Vis., 1–24 (2024)

    2024

  34. [42]

    In: ICCV (2023)

    Kong, L., Liu, Y., Li, X., Chen, R., Zhang, W., Ren, J., Pan, L., Chen, K., Liu, Z.: Robo3d: Towards robust and reliable 3d per- ception against corruptions. In: ICCV (2023)

    2023

  35. [43]

    Amer- ican Journal of Physics 53(10), 955–962 (1985)

    Drake, R., Gordon, J.: Mie scattering. Amer- ican Journal of Physics 53(10), 955–962 (1985)

    1985

  36. [44]

    Neural Networks 172, 106106 (2024)

    Hojjati, H., Ho, T.K.K., Armanfard, N.: Self- supervised anomaly detection in computer vision and beyond: A survey and outlook. Neural Networks 172, 106106 (2024)

    2024

  37. [45]

    Proceedings of the IEEE 109(5), 756– 795 (2021)

    Ruff, L., Kauffmann, J.R., Vandermeulen, R.A., Montavon, G., Samek, W., Kloft, M., Dietterich, T.G., M¨ uller, K.-R.: A unifying review of deep and shallow anomaly detec- tion. Proceedings of the IEEE 109(5), 756– 795 (2021)

    2021

  38. [46]

    In: ICCV (2017)

    Huang, X., Belongie, S.: Arbitrary style transfer in real-time with adaptive instance normalization. In: ICCV (2017)

    2017

  39. [47]

    In: CVPR (2025)

    Yang, L., Hu, P., Yuan, S., Zhang, L., Liu, J., Shen, H., Zhu, X.: Towards explicit geometry- reflectance collaboration for generalized lidar segmentation in adverse weather. In: CVPR (2025)

    2025

  40. [48]

    In: CVPR (2012)

    Geiger, A., Lenz, P., Urtasun, R.: Are we ready for autonomous driving? the kitti vision benchmark suite. In: CVPR (2012)

    2012

  41. [49]

    In: AAAI (2022) 15

    Xiao, A., Huang, J., Guan, D., Zhan, F., Lu, S.: Transfer learning from synthetic to real lidar point cloud for semantic segmentation. In: AAAI (2022) 15

    2022

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.