pith. machine review for the scientific record. sign in

arxiv: 2604.21387 · v1 · submitted 2026-04-23 · 💻 cs.CV

Recognition: unknown

EdgeFormer: local patch-based edge detection transformer on point clouds

Yifei Xie , Zhikun Tu , Tong Yang , Yuhe Zhang , Xinyu Zhou
Authors on Pith no claims yet

Pith reviewed 2026-05-09 22:36 UTC · model grok-4.3

classification 💻 cs.CV
keywords edge detectionpoint cloudstransformerlocal patches3D geometrypoint classificationfeature descriptors
0
0 comments X

The pith

EdgeFormer detects fine edges in point clouds by classifying points inside local patches.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper aims to solve the problem of detecting fine-grained edges on 3D point clouds, where edges are hard to find because they cluster densely or show only tiny surface changes. It does so by first building feature descriptors for the neighborhood around every point, then classifying whether each point lies on an edge. The conversion rests on the idea that nearby points correlate strongly enough to represent a local surface piece. Experiments report that the resulting EdgeFormer model matches or approaches the accuracy of six existing methods while handling finer details better than global approaches. This matters for any 3D vision task that needs reliable surface boundaries.

Core claim

EdgeFormer splits edge detection into two stages: it first creates local patch feature descriptors that summarize the neighborhood around each point, then feeds those descriptors into a transformer that classifies the point as edge or non-edge. The method therefore reduces the global edge-finding problem on the whole cloud to repeated local classification tasks.

What carries the argument

Local patch feature descriptors that summarize the neighborhood around each point and feed a transformer classifier.

If this is right

  • Fine details become accessible even when edges are densely packed or have only small gradients.
  • The full point cloud can be processed by handling one local patch at a time.
  • The same two-stage pipeline yields accuracy competitive with six prior edge detectors.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The patch-reduction idea could be tried on related point-cloud tasks such as normal estimation or segmentation.
  • Performance on real scanned data with varying density would test whether the local-correlation premise holds outside synthetic benchmarks.

Load-bearing premise

Spatially neighboring points exhibit high correlation and form a coherent local surface, so edge detection reduces to point classification inside those patches.

What would settle it

Run the model on point clouds deliberately sampled so that spatially neighboring points do not lie on smooth local surfaces, such as very sparse or randomly jittered clouds, and measure whether edge recall drops sharply.

read the original abstract

Edge points on 3D point clouds can clearly convey 3D geometry and surface characteristics, therefore, edge detection is widely used in many vision applications with high industrial and commercial demands. However, the fine-grained edge features are difficult to detect effectively as they are generally densely distributed or exhibit small-scale surface gradients. To address this issue, we present a learning-based edge detection network, named EdgeFormer, which mainly consists of two stages. Based on the observation that spatially neighboring points tend to exhibit high correlation, forming the local underlying surface, we convert the edge detection of the entire point cloud into a point classification based on local patches. Therefore, in the first stage, we construct local patch feature descriptors that describe the local neighborhood around each point. In the second stage, we classify each point by analyzing the local patch feature descriptors generated in the first stage. Due to the conversion of the point cloud into local patches, the proposed method can effectively extract the finer details. The experimental results show that our model demonstrates competitive performance compared to six baselines.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper proposes EdgeFormer, a two-stage transformer network for edge detection on 3D point clouds. It first constructs local patch feature descriptors around each point based on the assumption that spatially neighboring points form correlated local surfaces, then classifies each point as an edge or non-edge point by processing these descriptors. The central claim is that this local-patch conversion enables effective extraction of fine-grained edges and yields competitive performance against six unspecified baselines.

Significance. If the empirical claims are substantiated with proper datasets and metrics, the local-patch reduction could offer a practical way to handle dense or small-scale edges in point-cloud geometry processing, with potential utility in downstream tasks such as surface reconstruction or segmentation. The absence of any reported experiments, however, prevents assessment of whether the approach actually delivers the claimed advantage.

major comments (2)
  1. Abstract: the claim of 'competitive performance compared to six baselines' is presented without any datasets, quantitative metrics, error bars, ablation studies, or implementation details, rendering the central empirical contribution unverifiable from the manuscript text.
  2. Method description (first and second stages): the conversion of edge detection to point classification on local patches is motivated qualitatively but supplies no equations, network diagrams, or pseudocode for patch construction, feature descriptor computation, or the transformer classifier, blocking reproducibility and independent verification.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive review of our manuscript. We address each major comment point by point below and commit to revisions that will improve the clarity, verifiability, and reproducibility of the work.

read point-by-point responses

  1. Referee: Abstract: the claim of 'competitive performance compared to six baselines' is presented without any datasets, quantitative metrics, error bars, ablation studies, or implementation details, rendering the central empirical contribution unverifiable from the manuscript text.

    Authors: The referee correctly observes that the current manuscript text does not contain the supporting experimental details. We will add a full Experiments section reporting results on standard point-cloud datasets (including ABC and ShapeNet subsets), quantitative metrics (precision, recall, F1-score), error bars from multiple runs, ablation studies on patch size and transformer components, and implementation details such as training hyperparameters and baseline configurations. The abstract will be revised to briefly reference these elements so that the performance claim is immediately verifiable. revision: yes

  2. Referee: Method description (first and second stages): the conversion of edge detection to point classification on local patches is motivated qualitatively but supplies no equations, network diagrams, or pseudocode for patch construction, feature descriptor computation, or the transformer classifier, blocking reproducibility and independent verification.

    Authors: We agree that the method section requires formalization. In the revised manuscript we will introduce equations defining local patch extraction (neighborhood radius and point sampling), the computation of patch feature descriptors, and the transformer encoder-decoder used for per-point classification. A network architecture diagram will be added, together with pseudocode for the complete two-stage pipeline. These additions will enable independent re-implementation. revision: yes

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that local neighborhoods are highly correlated and sufficient for edge classification, plus the empirical effectiveness of the proposed architecture. No explicit free parameters or invented physical entities are introduced.

free parameters (1)
  • local patch size
    Design choice that determines the scale at which correlation is assumed; value not stated in abstract.
axioms (1)
  • domain assumption Spatially neighboring points tend to exhibit high correlation, forming the local underlying surface
    Explicitly invoked to justify converting edge detection into local-patch point classification.

pith-pipeline@v0.9.0 · 5489 in / 1266 out tokens · 136983 ms · 2026-05-09T22:36:41.583630+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 39 canonical work pages

  1. [1]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    Qi CR, Su H, Mo K, et al (2017) Pointnet: Deep learning on point sets for 3d classification and segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    2017

  2. [2]

    Computa- tional Visual Media 7:187–199

    Guo MH, Cai JX, Liu ZN, et al (2021) Pct: Point cloud transformer. Computa- tional Visual Media 7:187–199. https://doi.org/10.1007/s41095-021-0229-5

    work page doi:10.1007/s41095-021-0229-5 2021

  3. [3]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 8489–8499

    Tang L, Zhan Y, Chen Z, et al (2022) Contrastive boundary learning for point cloud segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 8489–8499

    2022

  4. [4]

    IEEE Trans Image Process 30:92–107

    Feng M, Gilani SZ, Wang Y, et al (2021) Relation graph network for 3d object detection in point clouds. IEEE Trans Image Process 30:92–107. https://doi.org/ 17 10.1109/TIP.2020.3031371

    work page doi:10.1109/tip.2020.3031371 2021

  5. [5]

    In: Proceedings of the International Conference on Machine Vision and Applications

    Wang L, Wang F, Yan F, et al (2018) Saliency-guide simplification for point-cloud geometry. In: Proceedings of the International Conference on Machine Vision and Applications. Association for Computing Machinery, New York, NY, USA, ICMVA 2018, pp 36–40, https://doi.org/10.1145/3220511.3220523

    work page doi:10.1145/3220511.3220523 2018

  6. [6]

    In: Proceedings of the 18th Meeting of the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games

    Zhang L, Sun Q, He Y (2014) Splatting lines: An efficient method for illus- trating 3d surfaces and volumes. In: Proceedings of the 18th Meeting of the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games. Asso- ciation for Computing Machinery, New York, NY, USA, I3D ’14, pp 135–142, https://doi.org/10.1145/2556700.2556703

    work page doi:10.1145/2556700.2556703 2014

  7. [7]

    ACM Trans Graph 40(5)

    Metzer G, Hanocka R, Giryes R, et al (2021) Self-sampling for neural point cloud consolidation. ACM Trans Graph 40(5). https://doi.org/10.1145/3470645

    work page doi:10.1145/3470645 2021

  8. [8]

    Multi- scale feature extraction on point-sampled surfaces

    Pauly M, Keiser R, Gross M (2003) Multi-scale feature extraction on point- sampled surfaces. Comput Graph Forum 22(3):281–289. https://doi.org/10.1111/ 1467-8659.00675

    work page arXiv 2003

  9. [9]

    IEEE Geosci Remote Sens Lett 14(8):1288–1292

    Xia S, Wang R (2017) A fast edge extraction method for mobile lidar point clouds. IEEE Geosci Remote Sens Lett 14(8):1288–1292. https://doi.org/10.1109/LGRS. 2017.2707467

    work page doi:10.1109/lgrs 2017

  10. [10]

    Sglbp: subgraph-based local binary patterns for feature extraction on point clouds

    Guo B, Zhang Y, Gao J, et al (2022) Sglbp: Subgraph-based local binary patterns for feature extraction on point clouds. Comput Graph Forum 41(6):51–66. https: //doi.org/10.1111/cgf.14500

    work page doi:10.1111/cgf.14500 2022

  11. [11]

    Vis Comput 24(6):449–462

    Daniels II J, Ochotta T, Ha LK, et al (2008) Spline-based feature curves from point-sampled geometry. Vis Comput 24(6):449–462. https://doi.org/10.1007/ s00371-008-0223-2

    2008

  12. [12]

    In: Computer Vision – ECCV 2018: 15th European Conference, Munich, Germany, September 8–14, 2018, Proceedings, Part VII

    Yu L, Li X, Fu CW, et al (2018) Ec-net: An edge-aware point set consolidation network. In: Computer Vision – ECCV 2018: 15th European Conference, Munich, Germany, September 8–14, 2018, Proceedings, Part VII. Springer-Verlag, Berlin, Heidelberg, pp 398–414, https://doi.org/10.1007/978-3-030-01234-2 24

    work page doi:10.1007/978-3-030-01234-2 2018

  13. [13]

    In: Proceedings of the 34th International Conference on Neural Information Processing Systems

    Wang X, Xu Y, Xu K, et al (2020) Pie-net: Parametric inference of point cloud edges. In: Proceedings of the 34th International Conference on Neural Information Processing Systems. Curran Associates Inc., Red Hook, NY, USA, NIPS’20

    2020

  14. [14]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 13601–13610

    Zhu X, Du D, Chen W, et al (2023) Nerve: Neural volumetric edges for parametric curve extraction from point cloud. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 13601–13610

    2023

  15. [15]

    3d is here: Point cloud library (pcl)

    Rusu RB, Cousins S (2011) 3d is here: Point cloud library (pcl). In: 2011 IEEE International Conference on Robotics and Automation. IEEE, Shanghai, China, 18 pp 1–4, https://doi.org/10.1109/ICRA.2011.5980567

    work page doi:10.1109/icra.2011.5980567 2011

  16. [16]

    A statistical approach for extraction of feature lines from point clouds

    Zhang Y, Geng G, Wei X, et al (2016) A statistical approach for extraction of feature lines from point clouds. Computers & Graphics 56:31–45. https://doi.org/ 10.1016/j.cag.2016.01.004

    work page doi:10.1016/j.cag.2016.01.004 2016

  17. [17]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Mo K, Zhu S, Chang AX, et al (2019) Partnet: A large-scale benchmark for fine- grained and hierarchical part-level 3d object understanding. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    2019

  18. [18]

    V oronoi-based curvature and feature estimation from point clouds

    M´ erigot Q, Ovsjanikov M, Guibas LJ (2011) Voronoi-based curvature and feature estimation from point clouds. IEEE Trans Vis Comput Graph 17(6):743–756. https://doi.org/10.1109/TVCG.2010.261

    work page doi:10.1109/tvcg.2010.261 2011

  19. [19]

    Comput Aided Des 39(4):276–283

    Demarsin K, Vanderstraeten D, Volodine T, et al (2007) Detection of closed sharp edges in point clouds using normal estimation and graph theory. Comput Aided Des 39(4):276–283. https://doi.org/10.1016/j.cad.2006.12.005

    work page doi:10.1016/j.cad.2006.12.005 2007

  20. [20]

    Fast and robust edge extraction in unorganized point clouds

    Bazazian D, Casas JR, Ruiz-Hidalgo J (2015) Fast and robust edge extraction in unorganized point clouds. In: 2015 International Conference on Digital Image Computing: Techniques and Applications (DICTA). IEEE, Adelaide, Australia, pp 1–8, https://doi.org/10.1109/DICTA.2015.7371262

    work page doi:10.1109/dicta.2015.7371262 2015

  21. [21]

    Proceedings of the 10th international meshing roundtable pp 293–305

    Gumhold S, Wang X, MacLeod R (2001) Feature extraction from point clouds. Proceedings of the 10th international meshing roundtable pp 293–305

    2001

  22. [22]

    In: Proceedings of the 2010 Shape Modeling International Conference

    Weber C, Hahmann S, Hagen H (2010) Sharp feature detection in point clouds. In: Proceedings of the 2010 Shape Modeling International Conference. IEEE Computer Society, USA, SMI ’10, pp 175–186, https://doi.org/10.1109/SMI.2010. 32

    work page doi:10.1109/smi.2010 2010

  23. [23]

    IEEE Trans Autom Sci Eng 19(2):1003–1016

    Chen H, Huang Y, Xie Q, et al (2022) Multiscale feature line extraction from raw point clouds based on local surface variation and anisotropic contraction. IEEE Trans Autom Sci Eng 19(2):1003–1016. https://doi.org/10.1109/TASE. 2021.3053006

    work page doi:10.1109/tase 2022

  24. [24]

    Robust smooth feature extraction from point clouds

    Daniels JI, Ha LK, Ochotta T, et al (2007) Robust smooth feature extraction from point clouds. In: Proceedings of the IEEE International Conference on Shape Modeling and Applications 2007. IEEE Computer Society, USA, SMI ’07, pp 123–136, https://doi.org/10.1109/SMI.2007.32

    work page doi:10.1109/smi.2007.32 2007

  25. [25]

    ACM Trans Graph 23(3):609–612

    Ohtake Y, Belyaev A, Seidel HP (2004) Ridge-valley lines on meshes via implicit surface fitting. ACM Trans Graph 23(3):609–612. https://doi.org/10. 1145/1015706.1015768

    work page arXiv 2004

  26. [26]

    ISPRS J Photogramm Remote Sens 102:172–183

    Lin Y, Wang C, Cheng J, et al (2015) Line segment extraction for large scale unorganized point clouds. ISPRS J Photogramm Remote Sens 102:172–183. https: 19 //doi.org/10.1016/j.isprsjprs.2014.12.027

    work page doi:10.1016/j.isprsjprs.2014.12.027 2015

  27. [27]

    Bako, Xinyi Liu, Leilani Battle, and Zhicheng Liu

    Stylianou G, Farin G (2004) Crest lines for surface segmentation and flattening. IEEE Trans Vis Comput Graph 10(5):536–544. https://doi.org/10.1109/TVCG. 2004.24

    work page doi:10.1109/tvcg 2004

  28. [28]

    In: Proceedings of the Third Eurographics Symposium on Geometry Processing

    Hildebrandt K, Polthier K, Wardetzky M (2005) Smooth feature lines on sur- face meshes. In: Proceedings of the Third Eurographics Symposium on Geometry Processing. Eurographics Association, Goslar, DEU, SGP ’05, pp 85–90

    2005

  29. [29]

    Computers & Graphics 82:222–231

    Gao Q, Yamaguchi Y (2019) Extraction of coherent and smooth feature lines from meshes with fine details. Computers & Graphics 82:222–231. https://doi.org/10. 1016/j.cag.2019.05.020

    2019

  30. [30]

    Autom Constr 101:59–71

    Wang Q, Sohn H, Cheng JC (2019) Development of high-accuracy edge line estimation algorithms using terrestrial laser scanning. Autom Constr 101:59–71. https://doi.org/10.1016/j.autcon.2019.01.009

    work page doi:10.1016/j.autcon.2019.01.009 2019

  31. [31]

    In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    Hackel T, Wegner JD, Schindler K (2016) Contour detection in unstructured 3d point clouds. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Las Vegas, NV, USA, pp 1610–1618, https://doi. org/10.1109/CVPR.2016.178

    work page doi:10.1109/cvpr.2016.178 2016

  32. [32]

    Höllt, N

    Loizou M, Averkiou M, Kalogerakis E (2020) Learning part boundaries from 3d point clouds. Comput Graph Forum 39(5):183–195. https://doi.org/10.1111/cgf. 14078

    work page doi:10.1111/cgf 2020

  33. [33]

    ACM Trans Graph 41(4)

    Matveev A, Rakhimov R, Artemov A, et al (2022) Def: Deep estimation of sharp geometric features in 3d shapes. ACM Trans Graph 41(4). https://doi.org/10. 1145/3528223.3530140

    work page arXiv 2022

  34. [34]

    In: Vedaldi A, Bischof H, Brox T, et al (eds) Computer Vision – ECCV 2020

    Hu Z, Zhen M, Bai X, et al (2020) Jsenet: Joint semantic segmentation and edge detection network for 3d point clouds. In: Vedaldi A, Bischof H, Brox T, et al (eds) Computer Vision – ECCV 2020. Springer International Publishing, Cham, pp 222–239

    2020

  35. [35]

    Edc-net: edge detection capsule network for 3d point clouds

    Bazazian D, Par´ es ME (2021) Edc-net: Edge detection capsule network for 3d point clouds. Applied Sciences 11(4):1833. https://doi.org/10.3390/app11041833

    work page doi:10.3390/app11041833 2021

  36. [36]

    Pcednet: a lightweight neural network for fast and interactive edge detection in 3d point clouds

    Himeur CE, Lejemble T, Pellegrini T, et al (2021) Pcednet: A lightweight neural network for fast and interactive edge detection in 3d point clouds. ACM Trans Graph 41(1). https://doi.org/10.1145/3481804

    work page doi:10.1145/3481804 2021

  37. [37]

    ISPRS J Photogramm Remote Sens 205:334–351

    Bode L, Weinmann M, Klein R (2023) Bounded: Neural boundary and edge detec- tion in 3d point clouds via local neighborhood statistics. ISPRS J Photogramm Remote Sens 205:334–351. https://doi.org/10.1016/j.isprsjprs.2023.09.023 20

    work page doi:10.1016/j.isprsjprs.2023.09.023 2023

  38. [38]

    Fast and power efficient GPU-based explicit elastic wave propagation analysis by low- ordered orthogonal voxel finite element with INT8 tensor cores

    Feng YF, Shen LY, Yuan CM, et al (2023) Deep shape representation with sharp feature preservation. Comput Aided Des 157:103468. https://doi.org/10.1016/j. cad.2022.103468

    work page doi:10.1016/j 2023

  39. [39]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 8486–8495

    Ye Y, Yi R, Gao Z, et al (2023) Nef: Neural edge fields for 3d parametric curve reconstruction from multi-view images. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 8486–8495

    2023

  40. [40]

    In: Guyon I, Luxburg UV, Bengio S, et al (eds) Advances in Neural Information Processing Systems

    Vaswani A, Shazeer N, Parmar N, et al (2017) Attention is all you need. In: Guyon I, Luxburg UV, Bengio S, et al (eds) Advances in Neural Information Processing Systems

    2017

  41. [41]

    IEEE trans Intell Transp Syst 23(12):24854–24865

    Lu D, Xie Q, Gao K, et al (2022) 3dctn: 3d convolution-transformer network for point cloud classification. IEEE trans Intell Transp Syst 23(12):24854–24865. https://doi.org/10.1109/TITS.2022.3198836

    work page doi:10.1109/tits.2022.3198836 2022

  42. [42]

    Khan, and Fahad Shah- baz Khan

    Zhao H, Jiang L, Jia J, et al (2021) Point transformer. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). IEEE, Mon- treal, QC, Canada, pp 16239–16248, https://doi.org/10.1109/ICCV48922.2021. 01595

    work page doi:10.1109/iccv48922.2021 2021

  43. [43]

    In: Proceedings of the Asian Conference on Computer Vision (ACCV)

    Du H, Yan X, Wang J, et al (2022) Point cloud upsampling via cascaded refinement network. In: Proceedings of the Asian Conference on Computer Vision (ACCV). Springer, Macao, China, pp 586–601, https://doi.org/10.1007/ 978-3-031-26319-4 7

    2022

  44. [44]

    Khan, and Fahad Shah- baz Khan

    Yu X, Rao Y, Wang Z, et al (2021) Pointr: Diverse point cloud completion with geometry-aware transformers. In: Proceedings of the IEEE/CVF Interna- tional Conference on Computer Vision (ICCV). IEEE, Montreal, QC, Canada, pp 12478–12487, https://doi.org/10.1109/ICCV48922.2021.01227

    work page doi:10.1109/iccv48922.2021.01227 2021

  45. [45]

    Journal of Graphics Tools 3(1):43–46

    Th¨ urrner G, W¨ uthrich CA (1998) Computing vertex normals from polygonal facets. Journal of Graphics Tools 3(1):43–46. https://doi.org/10.1080/10867651. 1998.10487487

    work page doi:10.1080/10867651 1998

  46. [46]

    Journal of Graphics Tools 4(2):1–6

    Max N (1999) Weights for computing vertex normals from facet normals. Journal of Graphics Tools 4(2):1–6. https://doi.org/10.1080/10867651.1999.10487501

    work page doi:10.1080/10867651.1999.10487501 1999

  47. [47]

    Abc: A big cad model dataset for geometric deep learning

    Koch S, Matveev A, Jiang Z, et al (2019) Abc: A big cad model dataset for geometric deep learning. In: Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR). IEEE, Long Beach, CA, USA, pp 9593–9603, https://doi.org/10.1109/CVPR.2019.00983

    work page doi:10.1109/cvpr.2019.00983 2019

  48. [48]

    In: Sensor Fusion IV: Control Paradigms and Data Structures, International Society for Optics and Photonics, pp 586–606, https://doi.org/10.1117/12.57955 21

    Besl PJ, McKay ND (1992) Method for registration of 3-D shapes. In: Sensor Fusion IV: Control Paradigms and Data Structures, International Society for Optics and Photonics, pp 586–606, https://doi.org/10.1117/12.57955 21

    work page doi:10.1117/12.57955 1992

  49. [49]

    In: 12th EG Workshop 3D Object Retrieval 2019, Italy, pp 1–8

    Moscoso Thompson E, Arvanitis G, Moustakas K, et al (2019) Shrec’19 track: Feature curve extraction on triangle meshes. In: 12th EG Workshop 3D Object Retrieval 2019, Italy, pp 1–8

    2019

  50. [50]

    Image and Vision Computing 16(9):635–651

    Johnson A, Hebert M (1998) Surface matching for object recognition in complex three-dimensional scenes. Image and Vision Computing 16(9):635–651. https:// doi.org/10.1016/S0262-8856(98)00074-2

    work page doi:10.1016/s0262-8856(98)00074-2 1998

  51. [51]

    In: Pajdla T, Matas J (eds) Computer Vision - ECCV 2004

    Frome A, Huber D, Kolluri R, et al (2004) Recognizing objects in range data using regional point descriptors. In: Pajdla T, Matas J (eds) Computer Vision - ECCV 2004. Springer Berlin Heidelberg, Berlin, Heidelberg, pp 224–237, https: //doi.org/10.1007/978-3-540-24672-5 18

    work page doi:10.1007/978-3-540-24672-5 2004

  52. [52]

    In: Daniilidis K, Maragos P, Paragios N (eds) Computer Vision – ECCV 2010

    Tombari F, Salti S, Di Stefano L (2010) Unique signatures of histograms for local surface description. In: Daniilidis K, Maragos P, Paragios N (eds) Computer Vision – ECCV 2010. Springer Berlin Heidelberg, Berlin, Heidelberg, pp 356–369, https://doi.org/10.1007/978-3-642-15558-1 26

    work page doi:10.1007/978-3-642-15558-1 2010

  53. [53]

    Roesthuis, and Sarthak Misra

    Rusu RB, Blodow N, Marton ZC, et al (2008) Aligning point cloud views using persistent feature histograms. In: 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp 3384–3391, https://doi.org/10.1109/IROS. 2008.4650967 22

    work page doi:10.1109/iros 2008