Pith. sign in

REVIEW 3 major objections 6 minor 65 references

LoDA: A Level of Detection Aware Method and a Multimodal Sensing Benchmark for Object Level Change Detection

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

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

Pith's one-line read By gating change decisions on a spatially varying level of detection, an object-level LiDAR change-detection pipeline beats the strongest baselines by 8.7 mIoU points on its own benchmark and transfers with frozen thresholds to Urb3DCD-V2.

desk verdict A detailed, honest object-level change-detection paper with a new benchmark and a sensible LoD-gating idea; the evaluation needs two reproducibility fixes before the headline margins are trustworthy. read the letter →

arxiv 2608.05356 v1 pith:LH5KKMQS submitted 2026-08-05 cs.CV cs.RO

Haitian Wang , Xinyu Wang , Sheldon Fung , Xian Zhang , Zichen Geng This is my paper · ORCID
classification cs.CVcs.RO
keywords LiDARchangedetectionobject-levellevelofmulti-temporalurbanmappingbenchmarkdatasetpointcloudregistrationinstancesegmentationmapupdating
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 argues that object-level change detection in multi-temporal urban LiDAR should be driven by an explicit, spatially varying detection limit rather than fixed thresholds or implicit learned priors. To that end, it builds a pipeline that fine-registers two vehicle-survey point clouds, computes a per-location Level of Detection (LoD95) from local roughness, sampling density, and pose uncertainty, and labels geometry-based, class-consistent instance pairs as Added, Removed, Increased, Decreased, or Unchanged using height, volume, and normal-displacement cues gated by LoD95. The paper also releases LoDA, an object-level benchmark built from 2023 and 2025 multimodal surveys of Subiaco, with fused maps, semantic and instance annotations, and five-class change labels. On LoDA the pipeline reports 95.0% accuracy, 90.8% macro F1, and 83.0% macro IoU, exceeding the strongest baseline by 8.7 IoU and 4.4 F1 points; with frozen thresholds it transfers to Urb3DCD-V2 at 96.81% mean accuracy and 89.52% mean change IoU. A reader should care because the output is a structured, audit-friendly map-update record, and because the interpretable knob—declare a change only if geometry moved more than the local sensor can detect—is where the paper locates the improvement.

What carries the argument

The central object is the spatially varying Level of Detection, $\mathrm{LoD}_{95}$, from Equation (2): a per-cell, normal-direction minimum detectable displacement that fuses local roughness (MAD-to-standard-deviation plane residuals), sampling density, and the translational block of the ICP pose covariance. It carries the argument because it converts a vague notion of observability into a quantitative gate: the change cues $\Delta h$, $\Delta V$, and $D_\perp$ are only trusted when they exceed $1.10\,\mathrm{LoD}_{95}$, and the pairwise detection limit is the median $\mathrm{LoD}_{95}$ over the overlap region. Around this gate the pipeline is organized as decoupled stages—Tukey-weighted registration, geometry-only proxy correspondence solved by Hungarian assignment, cut-pursuit superpoints with rule-based semantic and instance segmentation, and deterministic five-class decision rules—so that each stage's uncertainty is propagated forward instead of being absorbed by a learned black box.

What would settle it

Re-annotate a random sample of the LoDA cases that were excluded as ambiguous after QA, together with a stratified sample of grid cells with high LoD95, without letting annotators see any LoD map; then run the full pipeline and the strongest baseline on that sample. If the pipeline's mIoU margin over the baseline drops below the ablation's LoD-gating effect of 7.8 points, the benchmark construction rather than the method explains the result.

Watch

Extended reading notes

Core claim

At the paper's center is the claim that a change label is only warranted when the observed displacement exceeds what the sensing system could plausibly produce through noise, misalignment, or density variation. The mechanism is the LoD95 field of Equation (2), a per-cell normal-direction detection limit computed as $\mathrm{LoD}_{95}(g)=1.96\sqrt{\sigma^2_{23}(g)/N_{23}(g)+\sigma^2_{25}(g)/N_{25}(g)+n_g^{\top}\Sigma_t n_g}$, where the two variance terms come from robust plane-fit residuals in each epoch, the $N$'s are local sample counts, and $\Sigma_t$ is the translational block of the ICP pose covariance. Height differences $\Delta h$ and normal-direction displacements $D_\perp$ are only accepted when their magnitude exceeds $1.10\,\mathrm{LoD}_{95}$, and the volume and overlap statistics back the decision. The paper supports this claim with a controlled experiment: all thresholds are fixed once on the LoDA validation split, frozen for the test split and Urb3DCD-V2, and the ablation shows that removing the LoD gate costs 7.8 mIoU points, more than removing any single geometric cue. Stated at its strongest, the paper's discovery is that explicit observability modelling, not a larger network, is what separates reliable object-level change detection from per-point tile-based prediction.

Load-bearing premise

LoDA's ground-truth labels are assumed to be independent of the method's observability reasoning, but annotators used the LoD95 map as an observability aid and excluded ambiguous cases from benchmark statistics; if that aid and exclusion already suppress the same low-observability changes the pipeline is designed to ignore, the reported 8.7-point margin is partly a consequence of the benchmark construction.

Editorial extensions

If this is right

  • A map-update service can consume the output directly as structured records: each instance is assigned exactly one of Added, Removed, Increased, Decreased, or Unchanged, with a deterministic confidence score for ranking.
  • The method transfers across cities without per-dataset retuning: frozen thresholds yield 96.81% mAcc and 89.52% mIoUch on Urb3DCD-V2 under the official point-wise protocol, beating the strongest published baselines by 3.18 mIoUch points.
  • On LoDA, the largest gains are in the classes deep baselines handle worst: Increased and Decreased improve by 23.1 and 15.3 IoU points over the strongest baseline.
  • The ablation attributes the gains to the LoD gate: removing it cuts mIoU from 83.0 to 75.2, a larger drop than removing any single geometric cue such as height, volume, or normal displacement.
  • Because all thresholds are frozen, the benchmark enables controlled object-level evaluation without hidden per-dataset tuning; candidate instance pairs are generated automatically and final labels are QA-verified.

Reading between the lines

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

  • A direct test of the paper's main hypothesis would be to take a strong learning baseline and apply the same LoD gate to its point-wise predictions; if the margin on LoDA shrinks toward zero, the gate rather than the geometry pipeline is doing the work.
  • The LoD95 formula is generic enough to be dropped into other multi-temporal sensing settings, but the pose-covariance term would need to be replaced by the uncertainty of whatever trajectory solution is used; a useful extension is per-sensor calibration of the fixed sensor term.
  • The benchmark's annotation protocol had access to the LoD95 map as an observability aid, so a user should treat the reported margin as conditional on labels that may already be biased toward detectable changes; re-annotating the excluded ambiguous cases without the aid would settle this.
  • The per-block runtime scaling (47.6 to 75.8 seconds as points grow from 1.12 to 2.18 million) suggests the rule-based pipeline is practical at city-block scale, and the deterministic nature of all decisions makes it auditable for mapping products.
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 LoDA, a new object-level change detection benchmark for multitemporal vehicle LiDAR, and a three-stage pipeline that couples detection-limit-aware registration, geometric object proxies with rule-based semantic/instance segmentation, and LoD-gated height/volume/normal-displacement cues to assign five change labels (Added, Removed, Increased, Decreased, Unchanged). The method is evaluated on LoDA and, with frozen thresholds, on Urb3DCD-V2 under the official point-wise protocol, reporting strong improvements over published baselines (95.0% ACC, 90.8% mF1, 83.0% mIoU on LoDA; 96.81% mAcc, 89.52% mIoUch on Urb3DCD-V2).

Significance. If the results are reproducible, the paper makes a valuable contribution: it formalizes a level-of-detection notion for object-level change detection, releases a benchmark with explicit LoD information, and demonstrates cross-dataset transfer without retuning. The manuscript is notably detailed in reporting thresholds, ablations, and runtime, and the external Urb3DCD-V2 evaluation is a strong design choice. However, the unspecified baseline-to-object mapping on LoDA and the unspecified label projection on Urb3DCD-V2, together with the potential benchmark circularity from LoD-assisted annotation, leave the central quantitative claims insufficiently supported.

major comments (3)
  1. [Section 5.4 and Table 3, Panel B] Section 5.4 states that instance-level decisions are 'projected to point labels' on Urb3DCD-V2, but the projection from the five object-level labels (Added, Removed, Increased, Decreased, Unchanged) to the seven point-wise classes (Unchanged, New building, Demolition, New vegetation, Vegetation growth, Missing vegetation, Mobile) is never specified. Table 3 Panel B reports per-class IoUs for all seven classes, which implicitly requires a concrete mapping (e.g., how a building labeled 'Increased' is assigned, or how vegetation 'Decreased' maps to 'Missing vegetation' versus 'Unchanged'). Without this mapping, the reported 96.81% mAcc and 89.52% mIoUch are not reproducible, and the comparison against supervised point-wise baselines is not controlled. Please provide the exact label projection rule, specify whether it was fixed before seeing Urb3DCD-V2 labels, and state how unmatched object predictions are converted to point labels.
  2. [Section 3 (Data Collection)] The annotation protocol in Section 3 states that 'The local LoD95 map is available during annotation only as an observability aid' and that 'Cases that remain ambiguous after QA are excluded from the benchmark statistics rather than forced to follow the output of the proposed method.' This creates a potential circularity: if annotators relied on the LoD95 map when deciding that a low-observability change is indeterminate, and if those ambiguous cases are then excluded from evaluation, the benchmark's ground truth may encode the same suppression of low-observability changes that the method's LoD gating performs. The paper should report how many candidate instances were excluded at each stage, the reasons for exclusion, and ideally include a secondary evaluation that assigns a default label (e.g., Unchanged) to all excluded cases, so readers can assess whether the LoDA results are inflated by the exclusion. This is load-bearing because the LoDA benchmark is the primary vehicle for the central quantitative claim.
  3. [Section 5.1 and Tables 1/2] The LoDA evaluation compares object-based outputs against point-wise and voxel-wise baselines, but the protocol for converting baseline per-point predictions into the object-level five-class labels used in Tables 1 and 2 is not described. The paper does not state whether baseline predictions are aggregated by the same instance segmentation, by connected components, by voting, or by ground-truth instances, nor how a baseline that emits no object prediction is handled. Without this protocol, the reported 8.7 mIoU improvement over EFS KPConv is not a controlled comparison. Please specify the exact mapping from baseline point scores/classes to object predictions (including thresholds and grouping), or report baselines under their native point-wise protocol with an additional object-level aggregation step that is described in full.
minor comments (6)
  1. [Section 4.1, Eq. (2)] Equation (2) is presented as the definition of LoD95, but it is the M3C2-EP error propagation formula from [54]; please cite [54] at the equation and clarify which terms are adopted unchanged and which are estimated in this work.
  2. [Section 4.3] The sensitivity sweep for α_LoD over {1.0, 1.1, 1.2} reports that performance was stable, but no validation numbers are given; please report the validation mIoU for each α value.
  3. [Section 5.1] The statement that 'All rule thresholds ... were selected once on the LoDA validation split' should be reconciled with the many precise thresholds (e.g., 0.60 m, 0.12, 0.35 m, 0.18) that also appear in Section 4.3; please clarify whether any thresholds were taken from prior work or set by physical reasoning rather than the validation sweep.
  4. [Section 3] The paper says 'final labels are independently verified using synchronized annotation views and QA, reducing protocol bias,' but then states that the LoD95 map is used as an observability aid; please clarify the exact role of LoD95 in the verification step and how the annotators were instructed to use it.
  5. [Table 1] The reference numbering for EFS KPConv and DC3DCD is ambiguous because [20] is cited for both; please disambiguate the entries.
  6. [Section 5.2] The sentence 'Because compatible five-label object-level baselines are unavailable' should be supported by a short description of the adaptation protocol used for existing point-wise baselines, otherwise it is unclear whether the comparison is apples-to-apples.

Circularity Check

2 steps flagged · score 4.0 of 10

LoDA benchmark construction is partially self-referential: the ground truth is annotated with the LoD95 map as an 'observability aid' and QA-ambiguous cases are excluded, while the method's core is LoD gating; independent external transfer on Urb3DCD-V2 keeps the paper from being fully circular.

  1. other [Section 3, Data Collection, annotation protocol paragraph]
    "The local LoD95 map is available during annotation only as an observability aid and does not directly determine the released label. Cases that remain ambiguous after QA are excluded from the benchmark statistics rather than forced to follow the output of the proposed method."

    The headline result on LoDA is presented as evidence that LoD-gated object-level change detection works. But the benchmark labels are produced with the paper's LoD95 map visibly available to annotators as an 'observability aid,' and cases that remain ambiguous after QA are removed from the evaluation statistics. These are the low-observability cases that the method's LoD gating is specifically designed to suppress. The evaluation set is therefore conditioned on the same observability criterion that the method implements, so the reported 8.7 mIoU advantage over baselines is not a fully independent measurement of LoD gating's value: the benchmark is constructed, at least in part, around the method's central concept rather than testing it against an independently defined ground truth.

  2. other [Section 3 (instance definitions) versus Section 4.2 (instance extraction)]
    "Each epoch provides per-point semantic labels in {ground,building,vegetation,mobile} and per-class instance IDs defined on the fused map. Building instances are obtained by merging adjacent planar parts with compatible facade normals and overlapping footprints. Vegetation and mobile instances follow connectedness under a density-adaptive Euclidean radius."

    The benchmark's object-level units are generated by geometry rules that closely mirror the method's own instance formation in Section 4.2, where buildings are formed by merging planar patches with compatible normals, vegetation by density-adaptive Euclidean clustering, and mobile objects by horizontal clustering. The method is therefore evaluated on an object definition that is essentially the same as the one it uses internally. Its instance-level predictions are aligned with the benchmark's instance ground truth by construction, so the object-level scores do not measure the method's ability to generalize to independently defined objects or to compete with methods that segment instances under different criteria.

full rationale

The mathematical derivation chain itself is largely self-contained: the detection-limit formula in Eq. (2) is the external M3C2-EP error-propagation expression from Winiwarter et al. [54] with a pose-covariance term, not a quantity fitted to the paper's own labels. The pipeline thresholds, including alpha_LoD and semantic criteria, are selected once on the LoDA validation split and then frozen, so the LoDA test and Urb3DCD-V2 numbers are not fitted inputs. There is no load-bearing self-citation chain: the cited registration, M3C2, cut-pursuit, and matching methods are prior external work rather than the authors' own unverified claims. The main circularity is in the benchmark construction rather than in the equations. The LoDA ground truth is annotated with the LoD95 map available as an 'observability aid,' and QA-ambiguous cases are excluded from statistics; this makes the evaluation set partly conditioned on the paper's core observability concept, so the benchmark cannot independently test the method on the low-observability false-change cases that LoD gating targets. In addition, the benchmark instance IDs are produced by geometry rules closely matching the method's instance formation, making the object-level evaluation partially self-consistent. The Urb3DCD-V2 transfer is the strongest independent evidence, although the projection from the method's five object-level labels to Urb3DCD-V2's seven point-wise classes is never specified; that is a serious reproducibility omission but not itself a circularity. Overall the central claim retains independent content through the external benchmark, so a moderate partial-circularity score of 4 is appropriate rather than a higher score.

Assumptions & free parameters 10 free parameters · 5 assumptions · 0 invented entities

The pipeline introduces no new physical entities. Its quantitative results depend on a large set of hand-set thresholds (listed) and on the validity of prior uncertainty formulas. The most consequential modeling assumption is that the LoDA annotation protocol is not biased by LoD-aware observability reasoning.

free parameters (10)
  • alpha_LoD (LoD gate multiplier) = 1.10
    Selected by sensitivity sweep over {1.0, 1.1, 1.2} on LoDA validation; gates both height and normal-displacement cues.
  • Building increased/decreased height threshold = 0.60 m
    Threshold for |Delta h| in building pairs; fixed on validation.
  • Building increased/decreased volume ratio threshold = 0.12
    Relative volume change |Delta V|/V23 that flips a building pair to Increased/Decreased.
  • Vegetation increased/decreased height threshold = 0.35 m
    Threshold for vegetation height change, fixed on validation.
  • Vegetation increased/decreased volume ratio threshold = 0.18
    Relative volume change for vegetation, fixed on validation.
  • Correspondence candidate thresholds = centroid < 2.0 m, OBB IoU > 0.10
    Gate for geometry-proxy pairing before Hungarian assignment.
  • Hungarian unmatched cost = 3.0
    Controls Added/Removed vs. matched pairs in bipartite assignment.
  • Low-observability default LoD95 = 0.50 m
    Assigned when fewer than 30 points per epoch in a 1 m LoD cell; stops downstream change assertions.
  • Unchanged overlap and centroid bounds = IoU3D > 0.20, |Delta c| < 1.5 m (buildings)
    Defines the Unchanged class for matched building pairs; fixed on validation.
  • Semantic rule thresholds = P > 0.60, S > 0.35, mobile volume < 60 m^3, etc.
    Rule-based semantic segmentation thresholds selected on validation; propagate class-consistent instances into change cues.
assumptions (5)
  • domain assumption M3C2-EP error propagation formula (Eq. 2) is valid under Gaussian, independent residuals between epochs
    Equation (2) is lifted from Lague et al. 2013 and Winiwarter et al. 2021; the derivation relies on normality and independence of epoch errors.
  • domain assumption Residual rotational pose uncertainty is negligible after block-level rigid refinement
    Section 4.1 states the rotational term is neglected because it was 'empirically smaller than the local roughness term' on fixed 120 m blocks.
  • domain assumption Rule-based semantic and instance segmentation produces correct object correspondences for change analysis
    Section 4.2 uses hand-set geometry rules; if instances are mis-segmented, the object-level change labels inherit the error.
  • domain assumption The human-verified annotation protocol produces ground-truth labels independent of the method's inference rules
    Section 3: labels verified with synchronized plan-view renderings, height maps, occupancy overlays; LoD95 available as an aid; ambiguous cases excluded. Independence is assumed but not statistically demonstrated.
  • domain assumption Validation-chosen thresholds transfer to the test split and to Urb3DCD-V2
    Section 5.1 states all thresholds frozen after validation; this assumes the validation set is representative of test conditions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LoDA: A Level of Detection Aware Method and a Multimodal Sensing Benchmark for Object Level Change Detection." pith.science (2026). https://pith.science/paper/LH5KKMQS

@misc{pith2026260805356,
  author       = {Pith},
  title        = {Pith review of: LoDA: A Level of Detection Aware Method and a Multimodal Sensing Benchmark for Object Level Change Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LH5KKMQS}},
  note         = {Machine review of arXiv:2608.05356}
}
read the original abstract

High-definition 3D LiDAR maps are important for autonomous driving and smart-city services, which require reliable detection of object-level changes in multi-temporal urban LiDAR to keep digital maps aligned with the physical world. Existing approaches from raster height differencing to depth image and point-cloud networks often remain tile-based and threshold-driven, yielding per-point scores without explicit detection limits or consistent object-level labels. We propose an object-level 3D change-detection pipeline that integrates detection-limit-aware registration, geometry-driven object proxies with rule-based semantic and instance segmentation, and displacement cues in height, volume, and surface-normal direction to assign five change labels with confidence. By decoupling registration, geometry, and semantics, the pipeline propagates pose uncertainty into spatially varying detection limits, stabilizes cross-epoch correspondences, and suppresses false changes caused by residual misalignment and density variation. We also present LoDA, a level-of-detection (LoD) aware benchmark for the Subiaco district with fused multi-temporal vehicle-LiDAR maps constructed with LiDAR, GNSS, and IMU support, semantic instances, and object-level annotations. On this benchmark, our method achieves 95.0% accuracy, 90.8% macro F1, and 83.0% macro IoU, exceeding the best baseline by 8.7 IoU points and 4.4 F1 points. On the public Urb3DCD-V2 benchmark evaluated under the official point-wise protocol, it reaches 96.81% mean accuracy and 89.52% mean change IoU, improving over the strongest reported baselines by 1.36 points in mAcc and 3.18 points in mIoUch.

Figures

Figures reproduced from arXiv: 2608.05356 by the authors.

Figure 1
Figure 1. Overview of the full study. The left part summarises data acquisition and benchmark construction from raw LiDAR [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Acquisition layout in Subiaco. The survey is organised into 21 closed loops (Seq 1 to Seq 21) covering main corridors, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. LoD-aware object-level change-detection pipeline. Registration and correspondence produce LoD-aware matches, [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Representative geometry patterns used in the se [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Qualitative results on three Subiaco blocks. For [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

65 extracted references · 46 canonical work pages

  1. [1]

    Kaleab Taye Asrat and Hyung-Ju Cho. 2024. A comprehensive survey on high- definition map generation and maintenance.ISPRS International Journal of Geo-Information13, 7 (2024), 232

    work page 2024

  2. [2]

    Mohammad Awrangjeb, Clive S Fraser, and Guojun Lu. 2015. Building change detection from LiDAR point cloud data based on connected component analy- sis.ISPRS annals of the photogrammetry, remote sensing and spatial information sciences2 (2015), 393–400

    work page 2015

  3. [3]

    Jens Behley, Martin Garbade, Andres Milioto, Jan Quenzel, Sven Behnke, Cyrill Stachniss, and Jurgen Gall. 2019. Semantickitti: A dataset for semantic scene understanding of lidar sequences. InProceedings of the IEEE/CVF international conference on computer vision. 9297–9307

    work page 2019

  4. [4]

    Jon Louis Bentley. 1975. Multidimensional Binary Search Trees Used for Associa- tive Searching.Commun. ACM18, 9 (1975), 509–517. doi:10.1145/361002.361007

    arXiv 1975

  5. [5]

    Besl and Neil D

    Paul J. Besl and Neil D. McKay. 1992. A Method for Registration of 3-D Shapes. IEEE Transactions on Pattern Analysis and Machine Intelligence14, 2 (1992), 239–

    work page 1992

  6. [6]

    Peter Biber and Wolfgang Straßer. 2003. The normal distributions transform: A new approach to laser scan matching. InProceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003)(Cat. No. 03CH37453), Vol. 3. IEEE, 2743–2748

    work page 2003

  7. [7]

    Leo Breiman. 2001. Random Forests.Machine Learning45, 1 (2001), 5–32. doi:10. 1023/A:1010933404324

    work page 2001

  8. [8]

    Juan José Cabrera, Antonio Santo, Arturo Gil, Carlos Viegas, and Luis Payá. 2025. MinkUNeXt: Point cloud-based large-scale place recognition using 3D sparse convolutions.Array(2025), 100569

    work page 2025

Show all 65 references
  1. [9]

    Cesar Cadena, Luca Carlone, Henry Carrillo, Yasir Latif, Davide Scaramuzza, José Neira, Ian Reid, and John J Leonard. 2017. Past, present, and future of simultaneous localization and mapping: Toward the robust-perception age.IEEE Transactions on robotics32, 6 (2017), 1309–1332

    2017

  2. [10]

    Holger Caesar, Varun Bankiti, Alex H Lang, Sourabh Vora, Venice Erin Liong, Qiang Xu, Anush Krishnan, Yu Pan, Giancarlo Baldan, and Oscar Beijbom. 2020. nuscenes: A multimodal dataset for autonomous driving. InProceedings of the IEEE/CVF conference on computer vision and patte...

    2020

  3. [11]

    Andrea Censi. 2007. An accurate closed-form estimate of ICP’s covariance. In Proceedings 2007 IEEE international conference on robotics and automation. IEEE, 3167–3172

    2007

  4. [12]

    Xieyuanli Chen, Shijie Li, Benedikt Mersch, Louis Wiesmann, Jürgen Gall, Jens Behley, and Cyrill Stachniss. 2021. Moving object segmentation in 3D LiDAR data: A learning-based approach exploiting sequential data.IEEE Robotics and Automation Letters6, 4 (2021), 6529–6536

    2021

  5. [13]

    Yang Chen and Gérard Medioni. 1992. Object Modelling by Registration of Multiple Range Images.Image and Vision Computing10, 3 (1992), 145–155. doi:10.1016/0262-8856(92)90066-C

    1992 doi

  6. [14]

    Christopher Choy, JunYoung Gwak, and Silvio Savarese. 2019. 4d spatio-temporal convnets: Minkowski convolutional neural networks. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition. 3075–3084

    2019

  7. [15]

    Özgün Çiçek, Ahmed Abdulkadir, Soeren S Lienkamp, Thomas Brox, and Olaf Ronneberger. 2016. 3D U-Net: learning dense volumetric segmentation from sparse annotation. InInternational conference on medical image computing and computer-assisted intervention. Springer, 424–432

    2016

  8. [16]

    Rodrigo Caye Daudt, Bertr Le Saux, and Alexandre Boulch. 2018. Fully convo- lutional siamese networks for change detection. In2018 25th IEEE international conference on image processing (ICIP). IEEE, 4063–4067

    2018

  9. [17]

    Iris de Gélis, Zoé Bessin, Pierrick Letortu, Marie Jaud, Christophe Delacourt, Stéphane Costa, Olivier Maquaire, Rachel Davidson, Thomas Corpetti, and Sébastien Lefèvre. 2022. Cliff Change Detection Using Siamese KPConv Deep Network on 3D Point Clouds. InISPRS Annals of the Ph...

    2022 doi

  10. [18]

    Iris De Gélis, Thomas Corpetti, and Sébastien Lefèvre. 2024. Change detection needs change information: Improving deep 3-D point cloud change detection. IEEE Transactions on Geoscience and Remote Sensing62 (2024), 1–10

    2024

  11. [19]

    Iris de Gélis, Sébastien Lefèvre, and Thomas Corpetti. 2021. Change detection in urban point clouds: An experimental comparison with simulated 3d datasets. Remote Sensing13, 13 (2021), 2629

    2021

  12. [20]

    Iris de Gélis, Sébastien Lefèvre, and Thomas Corpetti. 2023. DC3DCD: Unsuper- vised learning for multiclass 3D point cloud change detection.ISPRS Journal of Photogrammetry and Remote Sensing206 (2023), 168–183

    2023

  13. [21]

    Martin Ester, Hans-Peter Kriegel, Jörg Sander, and Xiaowei Xu. 1996. A Density- Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. InProceedings of the Second International Conference on Knowledge Discovery and Data Mining. AAAI Press, 226–231

    1996

  14. [22]

    Martin A Fischler and Robert C Bolles. 1981. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography.Commun. ACM24, 6 (1981), 381–395

    1981

  15. [23]

    Benjamin Graham and Laurens Van der Maaten. 2017. Submanifold sparse convolutional networks.arXiv preprint arXiv:1706.01307(2017)

    2017 arXiv

  16. [24]

    Xu Han, Yuan Tang, Zhaoxuan Wang, and Xianzhi Li. 2024. Mamba3d: Enhancing local features for 3d point cloud analysis via state space model. InProceedings of the 32nd ACM International Conference on Multimedia. 4995–5004

    2024

  17. [25]

    Paul W Holland and Roy E Welsch. 1977. Robust regression using iteratively reweighted least-squares.Communications in Statistics-theory and Methods6, 9 (1977), 813–827

    1977

  18. [26]

    Xin Jin, Haisheng Su, Kai Liu, Cong Ma, Wei Wu, Fei Hui, and Junchi Yan. 2025. UniMamba: Unified spatial-channel representation learning with group-efficient mamba for LiDAR-based 3D object detection. InProceedings of the Computer Vision and Pattern Recognition Conference. 1407–1417

    2025

  19. [27]

    Abderrazzaq Kharroubi, Florent Poux, Zouhair Ballouch, Rafika Hajji, and Roland Billen. 2022. Three dimensional change detection using point clouds: A review. Geomatics2, 4 (2022), 457–485

    2022

  20. [28]

    Dimitri Lague, Nicolas Brodu, and Jérôme Leroux. 2013. Accurate 3D compar- ison of complex topography with terrestrial laser scanner: Application to the Rangitikei canyon (NZ).ISPRS journal of photogrammetry and remote sensing82 (2013), 10–26

    2013

  21. [29]

    Loic Landrieu and Guillaume Obozinski. 2017. Cut pursuit: Fast algorithms to learn piecewise constant functions on general weighted graphs.SIAM Journal on Imaging Sciences10, 4 (2017), 1724–1766

    2017

  22. [30]

    Loic Landrieu and Martin Simonovsky. 2018. Large-scale point cloud semantic segmentation with superpoint graphs. InProceedings of the IEEE conference on computer vision and pattern recognition. 4558–4567

    2018

  23. [31]

    Jesse Levinson, Michael Montemerlo, and Sebastian Thrun. 2007. Map-based precision vehicle localization in urban environments.. InRobotics: science and systems, Vol. 4. Atlanta, GA, USA, 1

    2007

  24. [32]

    Dingkang Liang, Xin Zhou, Wei Xu, Xingkui Zhu, Zhikang Zou, Xiaoqing Ye, Xiao Tan, and Xiang Bai. 2024. Pointmamba: A simple state space model for point cloud analysis.Advances in neural information processing systems37 (2024), 32653–32677

    2024

  25. [33]

    Ilya Loshchilov and Frank Hutter. 2017. SGDR: Stochastic Gradient Descent with Warm Restarts. InInternational Conference on Learning Representations. https://openreview.net/forum?id=Skq89Scxx

    2017

  26. [34]

    Ilya Loshchilov and Frank Hutter. 2019. Decoupled Weight Decay Regularization. InInternational Conference on Learning Representations. https://openreview.net/ forum?id=Bkg6RiCqY7

    2019

  27. [35]

    Jinhao Lu, Chenguang Dai, Zhenchao Zhang, Xuanguang Liu, Ruqin Zhou, Song Ji, Yongsheng Zhang, Haiyan Guan, and Hanyun Wang. 2025. Ms-DANet: Multi- scale Difference-aware Network for 3D Point Cloud Change Detection.IEEE Transactions on Geoscience and Remote Sensing(2025)

    2025

  28. [36]

    Martin Magnusson, Achim Lilienthal, and Tom Duckett. 2007. Scan registration for autonomous mining vehicles using 3D-NDT.Journal of Field Robotics24, 10 (2007), 803–827

    2007

  29. [37]

    James Munkres. 1957. Algorithms for the Assignment and Transportation Prob- lems.J. Soc. Indust. Appl. Math.5, 1 (1957), 32–38. doi:10.1137/0105003

    1957 doi

  30. [38]

    Hiroshi Murakami, Katsuto Nakagawa, Hiroyuki Hasegawa, Taku Shibata, and Eiji Iwanami. 1999. Change detection of buildings using an airborne laser scanner. ISPRS Journal of Photogrammetry and Remote Sensing54, 2-3 (1999), 148–152

    1999

  31. [39]

    Charles R Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas. 2017. Pointnet: Deep learning on point sets for 3d classification and segmentation. InProceedings of the IEEE conference on computer vision and pattern recognition. 652–660

    2017

  32. [40]

    Charles Ruizhongtai Qi, Li Yi, Hao Su, and Leonidas J Guibas. 2017. Pointnet++: Deep hierarchical feature learning on point sets in a metric space.Advances in neural information processing systems30 (2017)

    2017

  33. [41]

    Ayushman Ramola and Anurag Vidyarthi. 2024. Versatile applications of pixel- and object-based change detection techniques in remote sensing. In2024 15th international conference on computing communication and networking technologies (ICCCNT). IEEE, 1–6

    2024

  34. [42]

    Peter J Rousseeuw and Christophe Croux. 1993. Alternatives to the median absolute deviation.Journal of the American Statistical association88, 424 (1993), 1273–1283

    1993

  35. [43]

    Joseph Rowell, Lintong Zhang, and Maurice Fallon. 2024. LiSTA: Geometric object- based change detection in cluttered environments. In2024 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 3632–3638

    2024

  36. [44]

    Radu Bogdan Rusu and Steve Cousins. 2011. 3D Is Here: Point Cloud Library (PCL). In2011 IEEE International Conference on Robotics and Automation. IEEE, 1–4. doi:10.1109/ICRA.2011.5980567

    2011

  37. [45]

    Heiko G Seif and Xiaolong Hu. 2016. Autonomous driving in the iCity—HD maps as a key challenge of the automotive industry.Engineering2, 2 (2016), 159–162

    2016

  38. [46]

    Ayesha Shafique, Guo Cao, Zia Khan, Muhammad Asad, and Muhammad Aslam

  39. [47]

    Tixiao Shan, Brendan Englot, Drew Meyers, Wei Wang, Carlo Ratti, and Daniela Rus. 2020. Lio-sam: Tightly-coupled lidar inertial odometry via smoothing and mapping. In2020 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, 5135–5142. MM ’26, Novem...

    2020

  40. [48]

    2003.Morphological Image Analysis: Principles and Applications(2 ed.)

    Pierre Soille. 2003.Morphological Image Analysis: Principles and Applications(2 ed.). Springer, Berlin, Heidelberg. doi:10.1007/978-3-662-05088-0

    2003 doi

  41. [49]

    Pei Sun, Henrik Kretzschmar, Xerxes Dotiwalla, Aurelien Chouard, Vijaysai Patnaik, Paul Tsui, James Guo, Yin Zhou, Yuning Chai, Benjamin Caine, et al

  42. [50]

    Hugues Thomas, Charles R Qi, Jean-Emmanuel Deschaud, Beatriz Marcotegui, François Goulette, and Leonidas J Guibas. 2019. Kpconv: Flexible and deformable convolution for point clouds. InProceedings of the IEEE/CVF international confer- ence on computer vision. 6411–6420

    2019

  43. [51]

    Thi Huong Giang Tran, Camillo Ressl, and Norbert Pfeifer. 2018. Integrated change detection and classification in urban areas based on airborne laser scan- ning point clouds.Sensors18, 2 (2018), 448

    2018

  44. [52]

    Ignacio Vizzo, Tiziano Guadagnino, Benedikt Mersch, Louis Wiesmann, Jens Behley, and Cyrill Stachniss. 2023. Kiss-icp: In defense of point-to-point icp– simple, accurate, and robust registration if done the right way.IEEE Robotics and Automation Letters8, 2 (2023), 1029–1036

    2023

  45. [53]

    Tao Wang, Wei Wen, Jingzhi Zhai, Kang Xu, and Haoming Luo. 2024. Serialized point mamba: A serialized point cloud mamba segmentation model.arXiv preprint arXiv:2407.12319(2024)

    2024 arXiv

  46. [54]

    Lukas Winiwarter, Katharina Anders, and Bernhard Höfle. 2021. M3C2-EP: Pushing the limits of 3D topographic point cloud change detection by error propagation.ISPRS Journal of Photogrammetry and Remote Sensing178 (2021), 240–258

    2021

  47. [55]

    Xiaoyang Wu, Li Jiang, Peng-Shuai Wang, Zhijian Liu, Xihui Liu, Yu Qiao, Wanli Ouyang, Tong He, and Hengshuang Zhao. 2024. Point transformer v3: Simpler faster stronger. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition. 4840–4851

    2024

  48. [56]

    Wei Xu, Yixi Cai, Dongjiao He, Jiarong Lin, and Fu Zhang. 2022. Fast-lio2: Fast direct lidar-inertial odometry.IEEE Transactions on Robotics38, 4 (2022), 2053–2073

    2022

  49. [57]

    Yihui Yang and Volker Schwieger. 2023. Patch-based M3C2: Towards lower- uncertainty and higher-resolution deformation analysis of 3D point clouds.In- ternational Journal of Applied Earth Observation and Geoinformation125 (2023), 103535

    2023

  50. [58]

    Yuanwen Yue, Damien Robert, Jianyuan Wang, Sunghwan Hong, Jan Dirk Wegner, Christian Rupprecht, and Konrad Schindler. 2025. Litept: Lighter yet stronger point transformer.arXiv preprint arXiv:2512.13689(2025)

    2025

  51. [59]

    Wenxiao Zhan, Ruozhen Cheng, and Jing Chen. 2024. PGN3DCD: Prior- Knowledge-Guided Network for Urban 3-D Point Cloud Change Detection.IEEE Transactions on Geoscience and Remote Sensing62 (2024), 1–15

    2024

  52. [60]

    Guowen Zhang, Lue Fan, Chenhang He, Zhen Lei, Zhaoxiang Zhang, and Lei Zhang. 2024. Voxel mamba: Group-free state space models for point cloud based 3d object detection.Advances in Neural Information Processing Systems37 (2024), 81489–81509

    2024

  53. [61]

    Luqi Zhang, Haiping Wang, Chong Liu, Zhen Dong, and Bisheng Yang. 2026. ME-CPT: Multi-Task Enhanced Cross-Temporal Point Transformer for Urban 3D Change Detection.IEEE Transactions on Geoscience and Remote Sensing(2026)

    2026

  54. [62]

    Qian-Yi Zhou, Jaesik Park, and Vladlen Koltun. 2018. Open3D: A modern library for 3D data processing.arXiv preprint arXiv:1801.09847(2018)

    2018 arXiv

  55. [256]

    doi:10.1109/34.121791

    doi

  56. [2020]

    In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Scalability in perception for autonomous driving: Waymo open dataset. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2446–2454

  57. [2022]

    Remote Sensing14, 4 (2022), 871

    Deep learning-based change detection in remote sensing images: A review. Remote Sensing14, 4 (2022), 871

    2022

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.