Pith. sign in

REVIEW 4 major objections 6 minor 57 references

ESCAPE: Equivariant Shape Completion via Anchor Point Encoding

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read ESCAPE claims that a transformer over anchor-point distances, followed by coordinate optimization, completes partially scanned 3D shapes in arbitrary rotations without pose estimation.

desk verdict Useful empirical entry on rotation-equivariant completion, but the exact equivariance claim outruns the theory: coplanar anchors leave a reflection ambiguity the paper does not resolve. read the letter →

arxiv 2412.00952 v1 pith:IV2M36VH submitted 2024-12-01 cs.CV

Burak Bekci , Nassir Navab , Federico Tombari , Mahdi Saleh This is my paper · ORCID
classification cs.CV
keywords rotation-equivariantshapecompletionanchorpointencodingdistance-based3DrepresentationcloudtransformerarbitraryposefurthestsamplingChamferdistancereal-world
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

ESCAPE is a point-cloud completion method designed so that the answer is correct no matter how the partial input is rotated, without ever estimating the object's pose. It represents every input point by its distances to a small set of anchor points chosen from the cloud, feeds that rotation-invariant distance matrix to a transformer, predicts a distance matrix for the completed shape, and then solves for point coordinates. On rotated versions of the PCN benchmark the method reports Chamfer distance 10.58, while the strongest PCA-aligned baseline degrades to 26.65; on the OmniObject real-world benchmark it reports 18.82 versus 29.56 for the best comparison. The paper argues that the distance representation keeps reconstruction error constant with network depth and gives unique reconstructions once enough anchors are available. A sympathetic reader would take away that pose-free completion is achievable by replacing coordinates with anchor distances, at the price of slightly worse results on canonical inputs.

What carries the argument

The load-bearing object is the distance matrix $D \in \mathbb{R}^{n \times k}$ between the $n$ input points and $k = 8$ anchor points selected from the partial cloud by a deterministic farthest-point sampling initialized at the centroid, then refined within clusters to points of highest estimated curvature. Because distances are rotation-invariant while the anchor coordinates rotate with the input, a transformer that consumes only distances predicts a complete-shape distance matrix $\hat{D}_c$ that is also rotation-invariant; the final completed cloud is recovered by minimizing $\sum_j (\|p - a_j\|_2 - \hat{d}_{ij})^2$ per point with Levenberg-Marquardt, and the recovered coordinates inherit the input rotation. The paper backs this with a reconstruction-uniqueness theorem ($k \ge 4$ anchors in general position determine each point up to reflection) and an error-bound argument showing perturbations shift distances by at most the perturbation norm, independent of network depth.

What would settle it

Take a point cloud with an exact rotational symmetry, such as a uniform sphere or a cube sampled symmetrically, rotate it by an angle that permutes the equidistant candidates for one anchor, and compare the ESCAPE anchor set and completion for the two inputs; if the anchors or the completed clouds differ, the claimed exact rotation equivariance fails. A numerical version is to measure CD-L1 between completions of a symmetric input and its rotated copy across many random rotations and check whether every sample gives zero difference.

Watch

Extended reading notes

Core claim

The central claim is that rotation-equivariant shape completion can be achieved without specialized equivariant layers, pose estimation, or rotation augmentation, by encoding shapes as distances to rotation-covariant anchor points. The paper states that with its deterministic anchor selection and distance-based transformer, the final coordinates predicted by the pipeline retain the same orientation as the partial input cloud, making ESCAPE the only model where prediction is unaffected by the input rotation. Concretely, training only on canonical PCN data and testing on randomly rotated inputs yields CD-L1 of 8.14 to 13.86 per category, averaging 10.58, whereas PCA-aligned transformer baselines range from 26.65 to 92.15; the same weights transfer to unknown-pose OmniObject depth scans and rotated KITTI LiDAR cars. The paper also claims two theoretical properties of the distance representation: identical distance encodings imply isometric shapes, and input perturbations change the distance matrix by at most the perturbation size, a constant error bound independent of network depth.

Load-bearing premise

The whole equivariance chain rests on the anchor-selection procedure being exactly rotation-equivariant for every input, but the paper's deterministic farthest-point sampling does not specify how ties among equidistant points are broken; under rotation, a symmetric or near-symmetric cloud can therefore produce different anchors, different distance features, and a different completion.

Editorial extensions

If this is right

  • A completion model trained once on canonical data can be deployed on arbitrarily rotated scans without a pose-estimation preprocessing stage.
  • Because equivariance comes from the encoding rather than from custom layers, the same anchor-distance representation could be dropped into other point-cloud transformer architectures.
  • The constant error bound implies that deeper networks built on distance encodings should not accumulate rotational or positional error, unlike layerwise equivariant features.
  • The method accepts a canonical-input trade-off: existing aligned baselines score 6.53 to 8.38 on canonical PCN inputs while ESCAPE scores 10.58, meaning the equivariance guarantee is bought with some accuracy on perfectly aligned data.

Reading between the lines

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

  • Editorial inference: because the network sees only distances, the same ESCAPE weights can in principle complete shapes at varying levels of partiality without retraining, since the anchors adapt to whichever points are visible.
  • Editorial inference: the anchor-point mechanism is modular, so replacing curvature-based anchors with any rotation-equivariant keypoint detector is a direct extension that the paper's design already permits.
  • Editorial inference: the constant-error property suggests ESCAPE-style encodings could help other pose-free point-cloud tasks, such as registration or object detection, where deep equivariant networks currently suffer error accumulation.
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

4 major / 6 minor

Summary. ESCAPE proposes a rotation-equivariant point cloud completion pipeline. It selects k anchor points from the partial input via deterministic farthest point sampling followed by curvature-based refinement, represents all points by their Euclidean distances to the anchors (Eq. 1), predicts the distance matrix of the completed cloud with a transformer modified from AdaPoinTr, and finally recovers point coordinates by Levenberg-Marquardt optimization (Eq. 5). Experiments on PCN, OmniObject3D, and KITTI report large improvements over PCA-aligned and ConDor-canonicalized non-equivariant baselines on rotated inputs, together with ablations on anchor selection, noise robustness, and partiality. The appendix includes theoretical error-bound arguments, additional baseline comparisons, and a limitations section.

Significance. If the central equivariance claim were established exactly, ESCAPE would be a practically significant contribution: it avoids pose estimation modules and achieves a large empirical gap on rotated PCN inputs (CD-L1 10.58 vs. 26.65 for the best PCA-aligned baseline). The paper also provides useful baselines (SCARP, Vector Neurons, PPF-Snowflake, ConDor), real-world evaluations, and an ablation study on anchor selection. However, the current theoretical guarantees are overclaimed: the constant-error-bound theorem applies only to the distance map, not to the end-to-end completion, and the coordinate optimization has a reflection ambiguity for non-general-position anchors that breaks exact equivariance. The empirical results remain valuable, but the paper needs to either prove a qualified equivariance statement or substantially narrow the claim.

major comments (4)
  1. [Section 3.3, Eq. (5), and Section F] The end-to-end equivariance claim requires that the solution of Eq. (5) be rotation-equivariant. Theorem 1 guarantees uniqueness only when the anchors are in general position, but the anchor selection of Section 3.1 does not enforce this condition. For planar or near-planar partial scans (e.g., tables, desks, and many single-view depth maps), all k anchors can lie in or very near a single plane. In that regime the distances to the anchors determine each predicted point only up to reflection across the anchor plane, and the Levenberg-Marquardt initialization at the centroid of the anchors lies on that plane, where the normal-direction residual Jacobian is zero. The branch choice is therefore not coupled to the input rotation, so F(RP) = R F(P) is not guaranteed. Section F itself states that the optimization 'prevents the model from being rotation-invariant,' which contradicts the unqualified claim in Sections 3.3 and 4.2. Please add an explicit equivariance-error experiment on planar categories under in-plane rotations, and either prove an approximate-equivariance bound or restrict the claim to inputs whose selected anchors are in general position.
  2. [Appendix A.1, Eq. (10)] Theorem A.1 proves a Lipschitz bound on the input distance matrix, |d(p+epsilon,a)-d(p,a)| <= ||epsilon||, but it does not bound the error of the predicted completed point cloud. The end-to-end error additionally depends on the transformer's prediction error and on the optimization in Eq. (5). Calling this a constant error bound O(1) independent of network depth is therefore not supported for the full pipeline. In addition, the comparison with Vector Neurons in Eq. (8) is not parameter-free, since alpha depends on learned weight norms and on the rotation matrices; a fair comparison would require a bound on those norms or matched training conditions. Please state precisely which quantity enjoys the O(1) bound and which components of the pipeline are excluded.
  3. [Appendix B.1] Appendix B.1 claims that initializing FPS from the centroid gives 'exactly same results for the same input independent of rotation,' but no tie-breaking rule is specified. For inputs with symmetries, multiple points can be equidistant from the centroid or from the already-selected set, and a list-order tie-break is not rotation-equivariant; after rotation a different anchor set can be selected. The curvature refinement in Section 3.1 has a related ambiguity, because PCA-based curvature (Eqs. 3-4) is computed from eigendecompositions whose eigenvector signs and order are not canonically defined. Please specify a canonical tie-breaking rule, prove invariance for generic point clouds, and report a test on symmetric or near-symmetric shapes, since exact equivariance of the anchor selector is a load-bearing premise for the distance features.
  4. [Theorem 1 and Section 3.3] The theorem as stated says the distance matrix uniquely determines P up to rigid transformation, while Section 3.3 concedes the optimization has a unique solution only 'up to reflection.' These are different statements, since a reflection is not a rigid motion in SO(3), and the proof sketch's 'intersection of k spheres in general position yields a unique point' does not rigorously cover the n-point case or the coplanar-anchor case. Because reconstruction uniqueness is load-bearing for the coordinate optimization, please provide a complete proof with the precise general-position conditions and state clearly when only uniqueness up to reflection holds.
minor comments (6)
  1. [Section 4.2, Table 1] The claim that ESCAPE is 'the only model where prediction is unaffected by the input rotation' is based on a single random rotation; please report results over multiple rotation seeds, e.g., mean and standard deviation, to support the invariance statement empirically.
  2. [Section F] The phrase 'prevents the model from being rotation-invariant' should be distinguished from 'rotation-equivariant'; as written it appears to contradict Section 3.3 and should be clarified or reconciled.
  3. [Section 1] There is a typo in the contributions list: 'We present the an end-to-end rotation-equivariant shape completion method.'
  4. [References and Table 3] Reference [53] duplicates reference [52], and Table 3 uses 'MMID' where the metric is presumably MMD; please correct the notation.
  5. [Table 5 and Table 4] The ablation in Table 5 reports a CD-L1 of 14.74 for plain FPS on 'a subset of the PCN dataset,' while Table 4 reports an average of 10.58 for the full method; the subset size, composition, and relation to the main result should be specified.
  6. [Appendix A.1, Eqs. (7)-(8)] The symbol R is used both for a rotation matrix and for a layer output; please use distinct notation for clarity.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: equivariance is a constructive consequence of invariant distance features and covariant anchor coordinates, not a fitted or self-citational result.

full rationale

The derivation is self-contained and no step reduces to its own inputs. Rotation equivariance is obtained constructively: Eq. (1) defines distances that are invariant under a simultaneous rotation of points and anchors; the transformer consumes only these invariant distance features; and Eq. (5) is covariant because if p* minimizes the objective with anchors a_j, then Rp* minimizes the rotated objective with anchors Ra_j. Thus the output orientation is inherited from the input anchors by construction rather than from a fitted parameter relabeled as a prediction. Training uses direct supervision (DMCD, Eq. (6)) against ground-truth complete clouds on PCN, KITTI, and OmniObject, so the benchmark numbers are external evidence. The related-work citations to Riga and Rotation-Invariant Transformer are descriptive and not load-bearing; no uniqueness theorem is imported from the authors' prior work. Section F's limitation about the optimization 'prevent[ing] the model from being rotation-invariant' and the unspecified FPS tie-breaking in Appendix B.1 are correctness and robustness concerns about the equivariance guarantee, not circular reductions: they question whether the construction holds for all inputs, not whether the claim was assumed as an input.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The method rests on standard distance geometry facts and on the assumption that anchor selection remains rotation-equivariant under arbitrary rotations. No new physical entities are introduced. The main free parameters are the number of anchors (k=8) and the anchor refinement radius (0.075), both chosen by ablations.

free parameters (2)
  • number of anchor points k = 8
    Chosen as the 'optimal balance' from ablation Table 9; the reconstruction uniqueness theorem requires at least 4, and the method uses 8.
  • anchor refinement radius for curvature-based selection = 0.075
    The anchor selection ablation in Table 5 shows Ball Query radius 0.075 gives the best CD-L1 score; the main text does not explicitly fix this value, suggesting a hand-tuned design choice.
assumptions (5)
  • standard math A point in R^3 is uniquely determined by its distances to four or more non-coplanar anchor points in general position.
    Used in Theorem 1 and Section 3.3 to justify that the coordinate optimization has a unique solution.
  • domain assumption The selected anchors from the partial cloud are in general position, i.e., non-coplanar.
    The paper relies on k=8 FPS anchors being non-coplanar; no proof is given, and near-planar or symmetric subsets could violate this.
  • domain assumption Farthest point sampling with centroid initialization, together with curvature refinement, is exactly rotation-equivariant including tie-breaking.
    Appendix B.1 asserts determinism but does not specify a rotation-equivariant tie-breaking rule; exact equivariance can fail for symmetric inputs.
  • domain assumption A transformer trained only on canonical PCN data generalizes to arbitrary rotations because the input distance features are rotation-invariant.
    This is the empirical generalization assumption behind evaluating on rotated inputs without rotation augmentation; anchor selection consistency under rotation is a hidden part of it.
  • domain assumption The predicted distance matrix for the complete shape is realizable by some point set, so independent per-point optimization yields a coherent shape.
    Section 3.3 solves each point separately; if the transformer outputs inconsistent distances, the final cloud can contain distortions. The paper does not enforce global consistency.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ESCAPE: Equivariant Shape Completion via Anchor Point Encoding." pith.science (2026). https://pith.science/paper/IV2M36VH

@misc{pith2026241200952,
  author       = {Pith},
  title        = {Pith review of: ESCAPE: Equivariant Shape Completion via Anchor Point Encoding},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IV2M36VH}},
  note         = {Machine review of arXiv:2412.00952}
}
read the original abstract

Shape completion, a crucial task in 3D computer vision, involves predicting and filling the missing regions of scanned or partially observed objects. Current methods expect known pose or canonical coordinates and do not perform well under varying rotations, limiting their real-world applicability. We introduce ESCAPE (Equivariant Shape Completion via Anchor Point Encoding), a novel framework designed to achieve rotation-equivariant shape completion. Our approach employs a distinctive encoding strategy by selecting anchor points from a shape and representing all points as a distance to all anchor points. This enables the model to capture a consistent, rotation-equivariant understanding of the object's geometry. ESCAPE leverages a transformer architecture to encode and decode the distance transformations, ensuring that generated shape completions remain accurate and equivariant under rotational transformations. Subsequently, we perform optimization to calculate the predicted shapes from the encodings. Experimental evaluations demonstrate that ESCAPE achieves robust, high-quality reconstructions across arbitrary rotations and translations, showcasing its effectiveness in real-world applications without additional pose estimation modules.

Figures

Figures reproduced from arXiv: 2412.00952 by the authors.

Figure 1
Figure 1. Existing shape completion methods (Top) use conven [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall pipeline for ESCAPE model. Initially, we extract [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Qualitative comparison of models trained on PCN dataset and tested with rotated inputs. Each row contains the input to the model [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of models trained on PCN dataset [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparison of SCARP and ESCAPE on PCN [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Quantitative results of additive noise and point removal [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: Qualitative comparison of models fine-tuned on PCN [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 8
Figure 8. Figure 8: Heatmap of the point curvature values and derived anchor [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: Comparison of different anchor point selection algorithms. [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 51 canonical work pages

  1. [1]

    Wiley interdisciplinary reviews: computational statistics 2(4), 433– 459 (2010) 6

    Abdi, H., Williams, L.J.: Principal component analysis. Wiley interdisciplinary reviews: computational statistics 2(4), 433– 459 (2010) 6

    work page 2010

  2. [2]

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

    Ao, S., Hu, Q., Yang, B., Markham, A., Guo, Y .: Spinnet: Learning a general surface descriptor for 3d point cloud reg- istration. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 11753–11762 (2021) 2

    work page 2021

  3. [3]

    In: European Conference on Computer Vision

    Chen, Y ., Fernando, B., Bilen, H., Nießner, M., Gavves, E.: 3d equivariant graph implicit functions. In: European Conference on Computer Vision. pp. 485–502. Springer (2022) 3

    work page 2022

  4. [4]

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

    Chen, Z., Long, F., Qiu, Z., Yao, T., Zhou, W., Luo, J., Mei, T.: Anchorformer: Point cloud completion from discrimina- tive nodes. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 13581–13590 (2023) 3, 6, 7, 14

    work page 2023

  5. [5]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Deng, C., Litany, O., Duan, Y ., Poulenard, A., Tagliasacchi, A., Guibas, L.J.: Vector neurons: A general framework for so (3)-equivariant networks. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 12200– 12209 (2021) 2, 3, 8

    work page 2021

  6. [6]

    In: Proceedings of the European conference on computer vision (ECCV)

    Deng, H., Birdal, T., Ilic, S.: Ppf-foldnet: Unsupervised learn- ing of rotation invariant 3d local descriptors. In: Proceedings of the European conference on computer vision (ECCV). pp. 602–618 (2018) 2

    work page 2018

  7. [7]

    In: 2010 IEEE computer society conference on computer vision and pattern recognition

    Drost, B., Ulrich, M., Navab, N., Ilic, S.: Model globally, match locally: Efficient and robust 3d object recognition. In: 2010 IEEE computer society conference on computer vision and pattern recognition. pp. 998–1005. Ieee (2010) 2

    work page 2010

  8. [8]

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

    Fu, K., Liu, S., Luo, X., Wang, M.: Robust point cloud registration framework based on deep graph matching. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 8893–8902 (2021) 2

    work page 2021

Show all 57 references
  1. [9]

    In: Proc

    Geiger, A., Lenz, P., Urtasun, R.: Are we ready for Au- tonomous Driving? The KITTI Vision Benchmark Suite. In: Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). pp. 3354–3361 (2012) 5

    2012

  2. [10]

    Advances in neural information processing systems 27 (2014) 3

    Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde- Farley, D., Ozair, S., Courville, A., Bengio, Y .: Generative adversarial nets. Advances in neural information processing systems 27 (2014) 3

    2014

  3. [11]

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

    Graham, B., Engelcke, M., Van Der Maaten, L.: 3d seman- tic segmentation with submanifold sparse convolutional net- works. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 9224–9232 (2018) 3

    2018

  4. [12]

    arXiv preprint arXiv:2012.09688 (2020) 2, 3

    Guo, M.H., Cai, J.X., Liu, Z.N., Mu, T.J., Martin, R.R., Hu, S.M.: Pct: Point cloud transformer. arXiv preprint arXiv:2012.09688 (2020) 2, 3

    2020 arXiv

  5. [13]

    International journal of computer vision 105, 63–86 (2013) 2

    Guo, Y ., Sohel, F., Bennamoun, M., Lu, M., Wan, J.: Rota- tional projection statistics for 3d local surface description and object recognition. International journal of computer vision 105, 63–86 (2013) 2

    2013

  6. [14]

    In: Pro- ceedings of the 19th annual conference on computer graphics and interactive techniques

    Hoppe, H., DeRose, T., Duchamp, T., McDonald, J., Stuetzle, W.: Surface reconstruction from unorganized points. In: Pro- ceedings of the 19th annual conference on computer graphics and interactive techniques. pp. 71–78 (1992) 4

    1992

  7. [15]

    arXiv preprint arXiv:2103.03206 (2021) 2

    Jaegle, A., Gimeno, F., Brock, A., Zisserman, A., Vinyals, O., Carreira, J.: Perceiver: General perception with iterative attention. arXiv preprint arXiv:2103.03206 (2021) 2

    2021 arXiv

  8. [16]

    arXiv preprint arXiv:1312.6114 (2013) 3

    Kingma, D.P., Welling, M.: Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114 (2013) 3

    2013 arXiv

  9. [17]

    In: International Conference on Machine Learning

    Lee, J., Lee, Y ., Kim, J., Kosiorek, A., Choi, S., Teh, Y .W.: Set transformer: A framework for attention-based permutation- invariant neural networks. In: International Conference on Machine Learning. pp. 3744–3753. PMLR (2019) 2

    2019

  10. [18]

    Advances in neural information processing systems 31, 820–830 (2018) 2

    Li, Y ., Bu, R., Sun, M., Wu, W., Di, X., Chen, B.: Pointcnn: Convolution on x-transformed points. Advances in neural information processing systems 31, 820–830 (2018) 2

    2018

  11. [19]

    In: 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

    Maturana, D., Scherer, S.: V oxnet: A 3d convolutional neural network for real-time object recognition. In: 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). pp. 922–928. IEEE (2015) 2

    2015

  12. [20]

    In: Computer graphics forum

    Mitra, N.J., Pauly, M., Wand, M., Ceylan, D.: Symmetry in 3d geometry: Extraction and applications. In: Computer graphics forum. vol. 32, pp. 1–23. Wiley Online Library (2013) 3

    2013

  13. [21]

    In: Numerical analysis: proceedings of the biennial Conference held at Dundee, June 28–July 1, 1977

    Moré, J.J.: The levenberg-marquardt algorithm: implementa- tion and theory. In: Numerical analysis: proceedings of the biennial Conference held at Dundee, June 28–July 1, 1977. pp. 105–116. Springer (2006) 5

    2006

  14. [22]

    ACM Transactions on Graphics (TOG) 21(4), 807–832 (2002) 2

    Osada, R., Funkhouser, T., Chazelle, B., Dobkin, D.: Shape distributions. ACM Transactions on Graphics (TOG) 21(4), 807–832 (2002) 2

    2002

  15. [23]

    In: Symposium on ge- ometry processing

    Pauly, M., Mitra, N.J., Giesen, J., Gross, M.H., Guibas, L.J.: Example-based 3d scan completion. In: Symposium on ge- ometry processing. pp. 23–32 (2005) 3

    2005

  16. [24]

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

    Qi, C.R., Su, H., Mo, K., Guibas, L.J.: Pointnet: Deep learn- ing on point sets for 3d classification and segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 652–660 (2017) 1, 2, 3

    2017

  17. [25]

    arXiv preprint arXiv:1706.02413 (2017) 1, 2, 3

    Qi, C.R., Yi, L., Su, H., Guibas, L.J.: Pointnet++: Deep hierarchical feature learning on point sets in a metric space. arXiv preprint arXiv:1706.02413 (2017) 1, 2, 3

    2017 arXiv

  18. [26]

    arXiv preprint arXiv:1912.05905 (2019) 3

    Rosu, R.A., Schütt, P., Quenzel, J., Behnke, S.: Latticenet: Fast point cloud segmentation using permutohedral lattices. arXiv preprint arXiv:1912.05905 (2019) 3

    2019 arXiv

  19. [27]

    In: 2009 IEEE international conference on robotics and automation

    Rusu, R.B., Blodow, N., Beetz, M.: Fast point feature his- tograms (fpfh) for 3d registration. In: 2009 IEEE international conference on robotics and automation. pp. 3212–3217. IEEE (2009) 2

    2009

  20. [28]

    In: 2008 IEEE/RSJ international conference on intelligent robots and systems

    Rusu, R.B., Blodow, N., Marton, Z.C., Beetz, M.: Aligning point cloud views using persistent feature histograms. In: 2008 IEEE/RSJ international conference on intelligent robots and systems. pp. 3384–3391. IEEE (2008) 2

    2008

  21. [29]

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

    Sajnani, R., Poulenard, A., Jain, J., Dua, R., Guibas, L.J., Sridhar, S.: Condor: Self-supervised canonicalization of 3d pose for partial shapes. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 16969–16979 (2022) 13, 14

    2022

  22. [30]

    In: 2020 International Conference on 3D Vision (3DV)

    Saleh, M., Dehghani, S., Busam, B., Navab, N., Tombari, F.: Graphite: Graph-induced feature extraction for point cloud registration. In: 2020 International Conference on 3D Vision (3DV). pp. 241–251. IEEE (2020) 2

    2020

  23. [31]

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

    Saleh, M., Wu, S.C., Cosmo, L., Navab, N., Busam, B., Tombari, F.: Bending graphs: Hierarchical shape matching us- ing gated optimal transport. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11757–11767 (2022) 2

    2022

  24. [32]

    In: 2023 IEEE International Conference on Robotics and Automation (ICRA)

    Sen, B., Agarwal, A., Singh, G., Brojeshwar, B., Sridhar, S., Krishna, M.: Scarp: 3d shape completion in arbitrary poses for improved grasping. In: 2023 IEEE International Conference on Robotics and Automation (ICRA). pp. 3838–

    2023

  25. [33]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition

    Shi, R., Xue, Z., You, Y ., Lu, C.: Skeleton merger: an un- supervised aligned keypoint detector. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition. pp. 43–52 (2021) 12

    2021

  26. [34]

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

    Shi, S., Guo, C., Jiang, L., Wang, Z., Shi, J., Wang, X., Li, H.: Pv-rcnn: Point-voxel feature set abstraction for 3d object detection. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 10529–10538 (2020) 2

    2020

  27. [35]

    In: Computer Vision– ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part III 11

    Tombari, F., Salti, S., Di Stefano, L.: Unique signatures of histograms for local surface description. In: Computer Vision– ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part III 11. pp. 356–369. Springer (2010) 2

    2010

  28. [36]

    In: Advances in neural information processing systems

    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł., Polosukhin, I.: Attention is all you need. In: Advances in neural information processing systems. pp. 5998–6008 (2017) 2, 3

    2017

  29. [37]

    arXiv preprint arXiv:1710.10903 (2017) 2

    Veliˇckovi´c, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., Bengio, Y .: Graph attention networks. arXiv preprint arXiv:1710.10903 (2017) 2

    2017 arXiv

  30. [38]

    ACM Transactions On Graphics (TOG) 36(4), 1–11 (2017) 3

    Wang, P.S., Liu, Y ., Guo, Y .X., Sun, C.Y ., Tong, X.: O-cnn: Octree-based convolutional neural networks for 3d shape analysis. ACM Transactions On Graphics (TOG) 36(4), 1–11 (2017) 3

    2017

  31. [39]

    Acm Transactions On Graphics (tog) 38(5), 1–12 (2019) 2, 4

    Wang, Y ., Sun, Y ., Liu, Z., Sarma, S.E., Bronstein, M.M., Solomon, J.M.: Dynamic graph cnn for learning on point clouds. Acm Transactions On Graphics (tog) 38(5), 1–12 (2019) 2, 4

    2019

  32. [40]

    In: 2022 International Conference on 3D Vision (3DV)

    Wu, H., Miao, Y .: So (3) rotation equivariant point cloud completion using attention-based vector neurons. In: 2022 International Conference on 3D Vision (3DV). pp. 280–290. IEEE (2022) 3

    2022

  33. [41]

    In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2023) 2, 5, 6

    Wu, T., Zhang, J., Fu, X., Wang, Y ., Jiawei Ren, L.P., Wu, W., Yang, L., Wang, J., Qian, C., Lin, D., Liu, Z.: Omniobject3d: Large-vocabulary 3d object dataset for realistic perception, reconstruction and generation. In: IEEE/CVF Conference on Computer Vision and Pattern Reco...

    2023

  34. [42]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Xiang, P., Wen, X., Liu, Y .S., Cao, Y .P., Wan, P., Zheng, W., Han, Z.: Snowflakenet: Point cloud completion by snowflake point deconvolution with skip-transformer. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 5499–5509 (2021) 1, 3, 6, 7, 14

    2021

  35. [43]

    In: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

    Xu, C., Zhai, B., Wu, B., Li, T., Zhan, W., Vajda, P., Keutzer, K., Tomizuka, M.: You only group once: Efficient point-cloud processing with token representation and relation inference module. In: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)....

    2021

  36. [44]

    In: Proceed- ings of the IEEE conference on computer vision and pattern recognition

    Yang, Y ., Feng, C., Shen, Y ., Tian, D.: Foldingnet: Point cloud auto-encoder via deep grid deformation. In: Proceed- ings of the IEEE conference on computer vision and pattern recognition. pp. 206–215 (2018) 2

    2018

  37. [45]

    In: Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition

    Yew, Z.J., Lee, G.H.: Rpm-net: Robust point matching using learned features. In: Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition. pp. 11824– 11833 (2020) 2

    2020

  38. [46]

    IEEE Trans- actions on Pattern Analysis and Machine Intelligence (2024) 3

    Yu, H., Hou, J., Qin, Z., Saleh, M., Shugurov, I., Wang, K., Busam, B., Ilic, S.: Riga: Rotation-invariant and globally- aware descriptors for point cloud registration. IEEE Trans- actions on Pattern Analysis and Machine Intelligence (2024) 3

    2024

  39. [47]

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

    Yu, H., Qin, Z., Hou, J., Saleh, M., Li, D., Busam, B., Ilic, S.: Rotation-invariant transformer for point cloud matching. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 5384–5393 (2023) 3

    2023

  40. [48]

    In: Proceedings of the IEEE/CVF international con- ference on computer vision

    Yu, X., Rao, Y ., Wang, Z., Liu, Z., Lu, J., Zhou, J.: Pointr: Diverse point cloud completion with geometry-aware trans- formers. In: Proceedings of the IEEE/CVF international con- ference on computer vision. pp. 12498–12507 (2021) 3, 6, 7, 14

    2021

  41. [49]

    IEEE Transactions on Pattern Analysis and Machine Intelligence (2023) 3, 4, 6, 7, 14

    Yu, X., Rao, Y ., Wang, Z., Lu, J., Zhou, J.: Adapointr: Diverse point cloud completion with adaptive geometry-aware trans- formers. IEEE Transactions on Pattern Analysis and Machine Intelligence (2023) 3, 4, 6, 7, 14

    2023

  42. [50]

    In: 2018 international conference on 3D vision (3DV)

    Yuan, W., Khot, T., Held, D., Mertz, C., Hebert, M.: Pcn: Point completion network. In: 2018 international conference on 3D vision (3DV). pp. 728–737. IEEE (2018) 3, 5, 6

    2018

  43. [51]

    In: 2019 International conference on 3d vision (3DV)

    Zhang, Z., Hua, B.S., Rosen, D.W., Yeung, S.K.: Rotation invariant convolutions for 3d point clouds deep learning. In: 2019 International conference on 3d vision (3DV). pp. 204–

    2019

  44. [52]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Zhao, H., Jiang, L., Jia, J., Torr, P.H., Koltun, V .: Point transformer. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 16259–16268 (2021) 2

    2021

  45. [53]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Zhao, H., Jiang, L., Jia, J., Torr, P.H., Koltun, V .: Point transformer. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 16259–16268 (2021) 3

    2021

  46. [54]

    Advances in Neural Information Processing Systems 35, 7052–7064 (2022) 12

    Zhong, C., You, P., Chen, X., Zhao, H., Sun, F., Zhou, G., Mu, X., Gan, C., Huang, W.: Snake: Shape-aware neural 3d keypoint field. Advances in Neural Information Processing Systems 35, 7052–7064 (2022) 12

    2022

  47. [55]

    In: European conference on computer vision

    Zhou, H., Cao, Y ., Chu, W., Zhu, J., Lu, T., Tai, Y ., Wang, C.: Seedformer: Patch seeds based point cloud completion with upsample transformer. In: European conference on computer vision. pp. 416–432. Springer (2022) 1, 3, 6, 7, 13, 14

    2022

  48. [56]

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

    Zhou, Y ., Tuzel, O.: V oxelnet: End-to-end learning for point cloud based 3d object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 4490–4499 (2018) 2 A. Robustness Analysis A.1. Error Bounds This section provides detailed pro...

    2018

  49. [3845]

    IEEE (2023) 3, 6, 7, 8, 13

    2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.