Pith. sign in

REVIEW 1 major objections 5 minor 56 references

Imagine with the Teacher: Complete Shape in a Multi-View Distillation Way

T0 review · 1 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read By projecting partial scans into depth images and distilling a teacher that saw the complete shape, VD-PCN lowers the average PCN Chamfer distance to 6.32, below SVDFormer's 6.54, with parallel gains on ShapeNet-55 and MVP.

desk verdict Solid engineering report with impressive benchmark numbers, but the distillation mechanism is not actually isolated in the ablation, so the SOTA margin rests on a confounded intervention. read the letter →

arxiv 2501.19270 v1 pith:XONNYEVS submitted 2025-01-31 cs.CV

Zhanpeng Luo , Linna Wang , Guangwu Qian , Li Lu This is my paper · ORCID
classification cs.CV
keywords pointcloudcompletionknowledgedistillationmulti-viewlearningteacher-studenttrainingdepthimageprojectionfeaturealignmentU-NetencoderChamferdistance
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 sets out to show that point cloud completion — reconstructing the occluded part of an object from a partial 3D scan — is better handled as a 2D problem taught by a teacher. VD-PCN projects the partial cloud into six depth images, encodes them with a shared U-Net, and aggregates the views with attention before decoding back to a dense point cloud, all measured by Chamfer distance, the standard average point-to-point distance between predicted and ground-truth shapes. Completion knowledge is then transferred by a teacher–student scheme: a frozen teacher trained on complete depth maps supervises a student that sees only partial depth images, aligning their features on the pixel grid. The authors argue this pixel-grid alignment is what makes distillation work where point-feature distillation caused negative transfer. If the claim holds, a multi-view 2D encoder paired with feature-level distillation outperforms prior published methods on the three benchmarks, with the paper reporting averages that beat all compared baselines on PCN, ShapeNet-55, and MVP.

What carries the argument

The load-bearing object is the teacher–student distillation pair built on a multi-view perceptual encoder. Partial and complete point clouds are each projected into $k=6$ depth images at $224\times224$; a shared U-Net backbone (its left encoder branch) produces per-view feature maps, which an Intra-View Fusion cross-attention layer and an Intra-View Enhancement self-attention layer aggregate into a global feature tensor. The frozen teacher is trained on complete depth maps plus the partial cloud; the student sees only partial depth images, and its per-view feature map is aligned to the teacher's with an $\ell^2$ loss while the pooled global features are aligned with an $\ell^1$ loss. The named mechanism — 'the orderliness of pixels' — is what makes that alignment meaningful: because depth images form regular grids, teacher and student features at the same pixel position are directly comparable, sidestepping the correspondence problem that the paper says caused negative transfer in point-level distillation. A dual-modality decoder then reintroduces both 2D view features and 3D point features during coarse-to-fine upsampling.

What would settle it

A control experiment settles it: take the teacher-initialized student, remove both distillation losses ($f_v$ and $f_g$), and fine-tune on partial depth images alone; if the Chamfer distance stays near 6.32, the distillation signal itself is doing no work. A second check is to run the same two-stage training on non-ShapeNet data such as real LiDAR or RGB-D scans — the paper's own limitation note in Section 4.3 concedes all benchmarks are artificially derived from ShapeNet — and see whether the distillation margin survives out-of-distribution occlusion and noise.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is a training recipe rather than a new loss: complete depth maps shown to a teacher encode the 'imagination' that can be distilled into a student at the feature level, because both operate on regular pixel grids. The paper reports an average L1 Chamfer distance of $6.32\times10^{-3}$ on PCN, beating SVDFormer's $6.54\times10^{-3}$; an average L2 Chamfer distance of $0.70\times10^{-3}$ on ShapeNet-55 versus AnchorFormer's $0.76\times10^{-3}$; and $2.16\times10^{-4}$ on MVP. Its ablation (Table 6) attributes the gain to distillation: adding both per-view feature-map alignment ($f_v$, with $\ell^2$ loss) and pooled global-feature alignment ($f_g$, with $\ell^1$ loss) moves the model from 6.55 to 6.32. The central claim is that teacher knowledge learned from the complete depth map transfers through a shared U-Net encoder to the student's partial-view features, producing better coarse seed points and, from them, better final shapes.

Load-bearing premise

The entire improvement rests on one premise: that feature-level knowledge from a teacher trained on complete depth images transfers useful completion information to a student seeing only partial depth images, and that pixel-grid alignment between the two projected views is what makes that transfer work; if the alignment conveys nothing, the distillation component — the reported source of the main gain — contributes little.

Editorial extensions

If this is right

  • On the three benchmarks tested, a 2D multi-view encoder plus feature-level distillation beats every compared baseline, including the previous PCN leader, by Chamfer distance.
  • The teacher is discarded at inference, so the reported accuracy comes at the student's cost alone: 105.49 GFLOPs and 15.00 ms per input in the paper's measurements.
  • Distillation improves the coarse seed point set first, and the improvement propagates through upsampling to the final dense output, so the strategy can be grafted onto other coarse-to-fine decoders.
  • Because the encoder accepts depth images, input point resolution is not capped by point-based network width, a headroom the paper notes it does not use in its fair comparisons.
  • The pixel-aligned teacher–student scheme is presented as a reusable training diagram for other completion-like tasks, as stated in the conclusion.

Reading between the lines

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

  • A control that fine-tunes a teacher-initialized student with the distillation losses removed entirely would reveal how much of the 6.55-to-6.32 gain is the alignment signal itself rather than weight warm-starting, since the paper does not isolate these two factors.
  • Because the teacher saw complete depth maps and the student only partial ones, the distillation could be teaching hallucination of the missing region rather than geometric reasoning; feeding both networks identical partial views in a control experiment would separate these mechanisms.
  • If pixel alignment is truly the mechanism, the same recipe should transfer to tasks with the same correspondence failure, such as LiDAR scan completion, and it carries a testable prediction: the distillation gain should shrink as the number of views drops below the fixed six.
  • The paper's appended checklist (Section 'Experiment Statistical Significance') asserts that error bars were collected to verify statistical significance, but no error bars appear in the reported tables, so that supporting claim is not yet visible in the paper's own results.
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

1 major / 5 minor

Summary. The paper proposes VD-PCN, a point cloud completion network built around a multi-view 2D U-Net encoder, a cross-view attention aggregation module, a dual-modality decoder, and a teacher-student distillation strategy. The teacher is trained on complete depth maps together with partial point clouds, while the student observes only partial depth maps and is trained to mimic the teacher's intermediate feature maps under an L2/L1 distillation loss, with the student initialized from the teacher's weights. The method is evaluated on PCN, ShapeNet-55, and MVP, where the authors report state-of-the-art Chamfer distances (e.g., CD-L1 6.32 on PCN versus 6.54 for SVDFormer). The central mechanistic claim is that pixel-aligned feature distillation from a teacher that saw complete depth images is what enables the improvement, because pixel-grid alignment avoids the negative transfer seen in prior point-feature distillation approaches.

Significance. If the reported numbers are reproducible, the paper makes a useful empirical contribution: it shows that a relatively simple multi-view CNN encoder combined with teacher-student feature distillation can outperform more complex point-based completion architectures on standard benchmarks, and it offers a concrete explanation for why feature-level distillation succeeds in this setting when prior point-feature distillation failed. The paper is clearly written, the architecture is well motivated, and the benchmark coverage (PCN, ShapeNet-55, MVP) is broad. However, the significance is currently conditional: the main SOTA margin rests on a training intervention that is not cleanly isolated in the ablation study, and the absence of any reported variance makes it impossible to judge whether the headline differences are statistically meaningful. The code is only promised, not yet available, so independent verification is not currently possible.

major comments (1)
  1. [Section 4.1, NeurIPS Checklist item 7] The checklist states that the authors 'conduct multiple times of experiment and ablation study and provide the error bar to verify the statistical significance,' but no error bars, confidence intervals, or significance tests appear anywhere in the paper, including Tables 1-6. The reported differences are small (e.g., 6.32 vs. 6.54 CD-L1 on PCN, and similar-sized gaps on ShapeNet-55 and MVP), so run-to-run variance could plausibly change the ranking. The authors should report mean and standard deviation (or equivalent) over multiple seeds for at least the main tables and the ablation table, and state how many runs were used. This is necessary for the reader to assess whether the claimed improvements are real.
minor comments (5)
  1. [Section 3.2, Equation for LKD] The distillation loss equation uses P_T_v and P_S_v, but the text defines the aligned quantities as feature maps F_T_v and F_S_v; the notation should be made consistent. In addition, Section 4.1 says 'MSE loss for the teacher / student's global feature alignment and L1 loss for teacher / student's visual feature loss,' which appears reversed relative to Section 3.2, where F_v is aligned with L2 and F_g with L1. Please harmonize these descriptions.
  2. [Table 4] The caption of Table 4 says 'Comparison of methods on the MVP dataset,' but the table content (CD-L2 S/M/H, CD avg, F-Score) and the surrounding text indicate that this table reports ShapeNet-55 results. The caption should be corrected. In Table 2, the entry '00.62' for the Car category contains a typographical double zero.
  3. [Section 4.3, Figure 3] The text says 'We choose two easier categories and two harder categories' for the qualitative visualization, but does not identify which categories are considered easier and which harder; please specify or remove this claim.
  4. [Section 3.1] The paper says 'froze all module except the multi-view encoder and train it with a smaller learning rate,' but later describes training the student with both distillation losses and point-cloud reconstruction losses. Please clarify which modules are actually updated during the distillation stage, since this is important for reproducibility.
  5. [Table 4] The F-Score for Ours (0.558) is identical to AnchorFormer (0.558) on ShapeNet-55, so the claim of consistent superiority should be stated with this caveat rather than presented as an unqualified improvement on all metrics.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: VD-PCN's central claims are measured benchmark comparisons, and no predicted quantity reduces to an input by construction.

full rationale

This is an empirical paper with no theoretical derivation chain, so the main circularity patterns do not apply. The central claim is a measured Chamfer-distance improvement over external baselines on PCN, ShapeNet-55, and MVP; those numbers are benchmark evaluations, not quantities derived from fitted parameters. The distillation loss LKD is a training objective (Section 3.2, Eq. for LKD and Section 3.3), not a fitted constant renamed as a prediction. Hyperparameters such as k=6, Pc=128, and tau=1 are described as standard or empirically similar choices, not as values chosen to force the reported result. The only author-overlap citation, FBNet [18], appears in related work as an external baseline and is not load-bearing for any claim. The Table 6 ablation removes the entire knowledge-transfer bundle, including teacher-weight initialization, so the attribution of the 0.23 CD improvement specifically to pixel-aligned feature distillation is under-supported; however, that is an experimental-confound or attribution concern, not circularity, because the improvement is measured rather than entailed by the paper's definitions or equations. No equation equates an output with an input by construction, and no self-citation chain is invoked to rule out alternative methods. The checklist's claim that error bars are provided is not visibly supported by the tables, but that is a completeness/reproducibility issue, not a circularity issue.

Assumptions & free parameters 9 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard domain assumptions about multi-view projection and distillation transfer; no invented entities are introduced. The free parameters are conventional hyperparameters tuned on the benchmark validation sets.

free parameters (9)
  • Number of viewpoints k = 6
    Set by the authors; controls how much of the object is visible in the multi-view input. No sensitivity study is reported.
  • Depth image resolution = 224x224
    Chosen for the U-Net backbone; affects feature resolution.
  • Number of coarse points Pc = 128
    Selected because the authors 'find that set Pc to 128 or 256 would produce better performance, comparing to 512 and 1024' (Section 4.1).
  • Distillation loss scale factors tau0, tau1, tau2 = 1.0
    Set to 1 because 'different type of loss and scale factor tau1 and tau2 seem to achieve similar effect at the end' (Section 3.2).
  • Teacher learning rate = 0.0002
    Standard schedule for the 200-epoch teacher training.
  • Student learning rate = 0.0001
    Smaller LR for the distillation stage.
  • Training epochs = 200
    Fixed for both teacher and student.
  • Number of coarse-to-fine upsampling stages = 2
    Set as 'we take two stage for coarse to fine point generation'.
  • U-Net feature channels / downsampling = c=512, 16x downsample
    Architecture hyperparameters; give [H1,W1]=14x14 feature maps.
assumptions (4)
  • domain assumption Projecting a partial point cloud to six depth images preserves enough shape information for completion.
    Section 3.1: the model uses only depth images from k views as the 2D input; if projection loses critical geometry, the method fails.
  • domain assumption Pixel-aligned feature distillation from a teacher that sees complete depth maps transfers useful completion knowledge to a student that sees partial depth maps.
    Section 3.2 'Distilled Knowledge Transfer'; this is the core premise of the method.
  • domain assumption Chamfer distance is an adequate metric for completion quality.
    Used for training and evaluation (Sections 3.3 and 4.1); if the metric is poor, the reported gains may not reflect perceptual quality.
  • domain assumption Multi-view cross-attention (IVF/IVE) aggregates views without introducing alignment errors.
    Section 3.2 'Multi-View Perceptual Encoder'.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Imagine with the Teacher: Complete Shape in a Multi-View Distillation Way." pith.science (2026). https://pith.science/paper/XONNYEVS

@misc{pith2026250119270,
  author       = {Pith},
  title        = {Pith review of: Imagine with the Teacher: Complete Shape in a Multi-View Distillation Way},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XONNYEVS}},
  note         = {Machine review of arXiv:2501.19270}
}
read the original abstract

Point cloud completion aims to recover the completed 3D shape of an object from its partial observation caused by occlusion, sensor's limitation, noise, etc. When some key semantic information is lost in the incomplete point cloud, the neural network needs to infer the missing part based on the input information. Intuitively we would apply an autoencoder architecture to solve this kind of problem, which take the incomplete point cloud as input and is supervised by the ground truth. This process that develops model's imagination from incomplete shape to complete shape is done automatically in the latent space. But the knowledge for mapping from incomplete to complete still remains dark and could be further explored. Motivated by the knowledge distillation's teacher-student learning strategy, we design a knowledge transfer way for completing 3d shape. In this work, we propose a novel View Distillation Point Completion Network (VD-PCN), which solve the completion problem by a multi-view distillation way. The design methodology fully leverages the orderliness of 2d pixels, flexibleness of 2d processing and powerfulness of 2d network. Extensive evaluations on PCN, ShapeNet55/34, and MVP datasets confirm the effectiveness of our design and knowledge transfer strategy, both quantitatively and qualitatively. Committed to facilitate ongoing research, we will make our code publicly available.

Figures

Figures reproduced from arXiv: 2501.19270 by the authors.

Figure 1
Figure 1. Model Overview Multi-View 3d Understanding Early works in 3d shape learning usually adopt multi-view pro￾jection or 3D voxelization to transform the irregular point clouds into regular representations[22]. Before the invention of pointnet, we have to operate 2d/3d CNN on these regular representations. But recently, SimpleView[7] has drawn people’s attention again to the usage of 2d network in 3d understanding. Simpl… view at source ↗
Figure 2
Figure 2. Model detail illustrated for three modules in the decoder part. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Visualizartion on the PCN Dataset. shape with a multi-view projection way. SVDFormer utilize the combination of point information and image information while VD-PCN fully leverage the 2d information and the orderliness of 2d pixels. As indicated by the results, the knowledge transfer help our model predict better coarse point Pc. We also provide a qualitative analysis on PCN dataset [PITH_FULL_IMAGE:figures/full_fi… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

56 extracted references · 45 canonical work pages

  1. [1]

    Pcn: Point completion network

    Wentao Yuan, Tejas Khot, David Held, Christoph Mertz, and Martial Hebert. Pcn: Point completion network. In 2018 international conference on 3D vision (3DV), pages 728–737. IEEE, 2018

    work page 2018

  2. [2]

    Grnet: Gridding residual network for dense point cloud completion

    Haozhe Xie, Hongxun Yao, Shangchen Zhou, Jiageng Mao, Shengping Zhang, and Wenxiu Sun. Grnet: Gridding residual network for dense point cloud completion. In European Conference on Computer Vision, pages 365–381. Springer, 2020

    work page 2020

  3. [3]

    Pointr: Diverse point cloud completion with geometry-aware transformers

    Xumin Yu, Yongming Rao, Ziyi Wang, Zuyan Liu, Jiwen Lu, and Jie Zhou. Pointr: Diverse point cloud completion with geometry-aware transformers. In Proceedings of the IEEE/CVF international conference on computer vision, pages 12498–12507, 2021

    work page 2021

  4. [4]

    Snowflakenet: Point cloud completion by snowflake point deconvolution with skip-transformer

    Peng Xiang, Xin Wen, Yu-Shen Liu, Yan-Pei Cao, Pengfei Wan, Wen Zheng, and Zhizhong Han. Snowflakenet: Point cloud completion by snowflake point deconvolution with skip-transformer. In Proceedings of the IEEE/CVF international conference on computer vision, pages 5499–5509, 2021

    work page 2021

  5. [5]

    Anchorformer: Point cloud completion from discriminative nodes

    Zhikai Chen, Fuchen Long, Zhaofan Qiu, Ting Yao, Wengang Zhou, Jiebo Luo, and Tao Mei. Anchorformer: Point cloud completion from discriminative nodes. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 13581–13590, 2023

    work page 2023

  6. [6]

    Svdformer: Complementing point cloud via self-view augmentation and self-structure dual-generator

    Zhe Zhu, Honghua Chen, Xing He, Weiming Wang, Jing Qin, and Mingqiang Wei. Svdformer: Complementing point cloud via self-view augmentation and self-structure dual-generator. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 14508– 14518, 2023

    work page 2023

  7. [7]

    Revisiting point cloud shape classification with a simple and effective baseline

    Ankit Goyal, Hei Law, Bowei Liu, Alejandro Newell, and Jia Deng. Revisiting point cloud shape classification with a simple and effective baseline. In International Conference on Machine Learning, pages 3809–3820. PMLR, 2021

    work page 2021

  8. [8]

    Let images give you more: Point cloud cross-modal training for shape analysis

    Xu Yan, Heshen Zhan, Chaoda Zheng, Jiantao Gao, Ruimao Zhang, Shuguang Cui, and Zhen Li. Let images give you more: Point cloud cross-modal training for shape analysis. Advances in Neural Information Processing Systems, 35:32398–32411, 2022

    work page 2022

Show all 56 references
  1. [9]

    Pointnet++: Deep hierarchical feature learning on point sets in a metric space

    Charles Ruizhongtai Qi, Li Yi, Hao Su, and Leonidas J Guibas. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in neural information processing systems, 30, 2017

    2017

  2. [10]

    Seedformer: Patch seeds based point cloud completion with upsample transformer

    Haoran Zhou, Yun Cao, Wenqing Chu, Junwei Zhu, Tong Lu, Ying Tai, and Chengjie Wang. Seedformer: Patch seeds based point cloud completion with upsample transformer. In European conference on computer vision, pages 416–432. Springer, 2022

    2022

  3. [11]

    High-fidelity point cloud completion with low-resolution recovery and noise-aware upsampling

    Ren-Wu Li, Bo Wang, Lin Gao, Ling-Xiao Zhang, and Chun-Peng Li. High-fidelity point cloud completion with low-resolution recovery and noise-aware upsampling. Graphical Models, 126: 101173, 2023

    2023

  4. [12]

    Csdn: Cross-modal shape-transfer dual-refinement network for point cloud completion

    Zhe Zhu, Liangliang Nan, Haoran Xie, Honghua Chen, Jun Wang, Mingqiang Wei, and Jing Qin. Csdn: Cross-modal shape-transfer dual-refinement network for point cloud completion. IEEE Transactions on Visualization and Computer Graphics, 2023

    2023

  5. [13]

    Cascaded refinement network for point cloud completion

    Xiaogang Wang, Marcelo H Ang Jr, and Gim Hee Lee. Cascaded refinement network for point cloud completion. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 790–799, 2020

    2020

  6. [14]

    Scancomplete: Large-scale scene completion and semantic segmentation for 3d scans

    Angela Dai, Daniel Ritchie, Martin Bokeloh, Scott Reed, Jürgen Sturm, and Matthias Nießner. Scancomplete: Large-scale scene completion and semantic segmentation for 3d scans. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 4578–4587, 2018

    2018

  7. [15]

    View-guided point cloud completion

    Xuancheng Zhang, Yutong Feng, Siqi Li, Changqing Zou, Hai Wan, Xibin Zhao, Yandong Guo, and Yue Gao. View-guided point cloud completion. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 15890–15899, 2021. 10

    2021

  8. [16]

    Cross-modal learning for image-guided point cloud shape completion

    Emanuele Aiello, Diego Valsesia, and Enrico Magli. Cross-modal learning for image-guided point cloud shape completion. Advances in Neural Information Processing Systems, 35:37349– 37362, 2022

    2022

  9. [17]

    Point cloud completion with pretrained text-to-image diffusion models

    Yoni Kasten, Ohad Rahamim, and Gal Chechik. Point cloud completion with pretrained text-to-image diffusion models. Advances in Neural Information Processing Systems, 36, 2024

    2024

  10. [18]

    Fbnet: Feedback network for point cloud completion

    Xuejun Yan, Hongyu Yan, Jingjing Wang, Hang Du, Zhihong Wu, Di Xie, Shiliang Pu, and Li Lu. Fbnet: Feedback network for point cloud completion. In European Conference on Computer Vision, pages 676–693. Springer, 2022

    2022

  11. [19]

    Lake-net: Topology-aware point cloud completion by localizing aligned keypoints

    Junshu Tang, Zhijun Gong, Ran Yi, Yuan Xie, and Lizhuang Ma. Lake-net: Topology-aware point cloud completion by localizing aligned keypoints. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1726–1735, 2022

    2022

  12. [20]

    Infocd: A contrastive chamfer distance loss for point cloud completion

    Fangzhou Lin, Yun Yue, Ziming Zhang, Songlin Hou, Kazunori Yamada, Vijaya Kolacha- lama, and Venkatesh Saligrama. Infocd: A contrastive chamfer distance loss for point cloud completion. Advances in Neural Information Processing Systems, 36, 2024

    2024

  13. [21]

    Hyperbolic chamfer distance for point cloud completion

    Fangzhou Lin, Yun Yue, Songlin Hou, Xuechu Yu, Yajun Xu, Kazunori D Yamada, and Ziming Zhang. Hyperbolic chamfer distance for point cloud completion. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 14595–14606, 2023

    2023

  14. [22]

    V olumetric and multi-view cnns for object classification on 3d data

    Charles R Qi, Hao Su, Matthias Nießner, Angela Dai, Mengyuan Yan, and Leonidas J Guibas. V olumetric and multi-view cnns for object classification on 3d data. InProceedings of the IEEE conference on computer vision and pattern recognition, pages 5648–5656, 2016

    2016

  15. [23]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part ...

    2015

  16. [24]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv...

    2010 arXiv

  17. [25]

    Imagenet: A large- scale hierarchical image database

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large- scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee, 2009

    2009

  18. [26]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020

    2020

  19. [27]

    Distilling the knowledge in a neural network

    Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531, 2015

    2015 arXiv

  20. [28]

    Pointdistiller: structured knowledge distillation towards efficient and compact 3d detection

    Linfeng Zhang, Runpei Dong, Hung-Shuo Tai, and Kaisheng Ma. Pointdistiller: structured knowledge distillation towards efficient and compact 3d detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 21791–21801, 2023

    2023

  21. [29]

    Prototransfer: Cross-modal prototype transfer for point cloud segmentation

    Pin Tang, Hai-Ming Xu, and Chao Ma. Prototransfer: Cross-modal prototype transfer for point cloud segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3337–3347, 2023

    2023

  22. [30]

    Leap: Liberate sparse-view 3d modeling from camera poses

    Hanwen Jiang, Zhenyu Jiang, Yue Zhao, and Qixing Huang. Leap: Liberate sparse-view 3d modeling from camera poses. arXiv preprint arXiv:2310.01410, 2023

    2023 arXiv

  23. [31]

    Fsc: Few-point shape completion

    Xianzu Wu, Xianfeng Wu, Tianyu Luan, Yajing Bai, Zhongyuan Lai, and Junsong Yuan. Fsc: Few-point shape completion. arXiv preprint arXiv:2403.07359, 2024

    2024 arXiv

  24. [32]

    Online knowledge distillation via collaborative learning

    Qiushan Guo, Xinjiang Wang, Yichao Wu, Zhipeng Yu, Ding Liang, Xiaolin Hu, and Ping Luo. Online knowledge distillation via collaborative learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11020–11029, 2020. 11

    2020

  25. [33]

    Variational relational point completion network

    Liang Pan, Xinyi Chen, Zhongang Cai, Junzhe Zhang, Haiyu Zhao, Shuai Yi, and Ziwei Liu. Variational relational point completion network. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8524–8533, 2021

    2021

  26. [34]

    Pytorch: An imperative style, high-performance deep learning library

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems, 32, 2019

    2019

  27. [35]

    Decoupled weight decay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017

    2017 arXiv

  28. [36]

    Foldingnet: Point cloud auto-encoder via deep grid deformation

    Yaoqing Yang, Chen Feng, Yiru Shen, and Dong Tian. Foldingnet: Point cloud auto-encoder via deep grid deformation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 206–215, 2018

    2018

  29. [37]

    Topnet: Structural point cloud decoder

    Lyne P Tchapmi, Vineet Kosaraju, Hamid Rezatofighi, Ian Reid, and Silvio Savarese. Topnet: Structural point cloud decoder. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 383–392, 2019

    2019

  30. [38]

    Morphing and sampling network for dense point cloud completion

    Minghua Liu, Lu Sheng, Sheng Yang, Jing Shao, and Shi-Min Hu. Morphing and sampling network for dense point cloud completion. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 11596–11603, 2020

    2020

  31. [39]

    Ecg: Edge-aware point cloud completion with graph convolution

    Liang Pan. Ecg: Edge-aware point cloud completion with graph convolution. IEEE Robotics and Automation Letters, 5(3):4392–4398, 2020

    2020

  32. [40]

    Cra-pcn: Point cloud completion with intra-and inter-level cross-resolution transformers

    Yi Rong, Haoran Zhou, Lixin Yuan, Cheng Mei, Jiahao Wang, and Tong Lu. Cra-pcn: Point cloud completion with intra-and inter-level cross-resolution transformers. arXiv preprint arXiv:2401.01552, 2024

    2024 arXiv

  33. [41]

    Shapenet: An information-rich 3d model repository

    Angel X Chang, Thomas Funkhouser, Leonidas Guibas, Pat Hanrahan, Qixing Huang, Zimo Li, Silvio Savarese, Manolis Savva, Shuran Song, Hao Su, et al. Shapenet: An information-rich 3d model repository. arXiv preprint arXiv:1512.03012, 2015. 12 NeurIPS Paper Checklist

    2015 arXiv

  34. [42]

    Guidelines: • The answer NA means that the abstract and introduction do not include the claims made in the paper

    Claims Question: Do the main claims made in the abstract and introduction accurately reflect the paper’s contributions and scope? Answer: [Yes] Justification: In the abstract and introduction, we share two observation in the completion task and give out solution. Guidelines: •...

  35. [43]

    Limitations

    Limitations Question: Does the paper discuss the limitations of the work performed by the authors? Answer: [Yes] Justification: In the 4.3 Analysis of model part, we give an discussion about the limitation of current work. Guidelines: • The answer NA means that the paper has n...

  36. [44]

    Guidelines: • The answer NA means that the paper does not include theoretical results

    Theory Assumptions and Proofs Question: For each theoretical result, does the paper provide the full set of assumptions and a complete (and correct) proof? 13 Answer: [No] Justification: There is no theoretical result in this paper. Guidelines: • The answer NA means that the p...

  37. [45]

    Guidelines: • The answer NA means that the paper does not include experiments

    Experimental Result Reproducibility Question: Does the paper fully disclose all the information needed to reproduce the main ex- perimental results of the paper to the extent that it affects the main claims and/or conclusions of the paper (regardless of whether the code and da...

  38. [46]

    Guidelines: • The answer NA means that paper does not include experiments requiring code

    Open access to data and code 14 Question: Does the paper provide open access to the data and code, with sufficient instruc- tions to faithfully reproduce the main experimental results, as described in supplemental material? Answer: [Yes] Justification: Yes, we would provide op...

  39. [47]

    Guidelines: • The answer NA means that the paper does not include experiments

    Experimental Setting/Details Question: Does the paper specify all the training and test details (e.g., data splits, hyper- parameters, how they were chosen, type of optimizer, etc.) necessary to understand the results? Answer: [Yes] Justification: We have sepcified all the tra...

  40. [48]

    Guidelines: • The answer NA means that the paper does not include experiments

    Experiment Statistical Significance Question: Does the paper report error bars suitably and correctly defined or other appropriate information about the statistical significance of the experiments? Answer: [Yes] Justification: We conduct multiple times of experiment and ablati...

  41. [49]

    But 4* Nvidia 3090 GPU can reproduce the whole experiment

    Experiments Compute Resources Question: For each experiment, does the paper provide sufficient information on the com- puter resources (type of compute workers, memory, time of execution) needed to reproduce the experiments? Answer: [Yes] Justification: For the experiments, we...

  42. [50]

    Guidelines: • The answer NA means that the authors have not reviewed the NeurIPS Code of Ethics

    Code Of Ethics Question: Does the research conducted in the paper conform, in every respect, with the NeurIPS Code of Ethics https://neurips.cc/public/EthicsGuidelines? Answer: [Yes] Justification: We have checked the NeurIPS code of Ethics and the whole research was the resea...

  43. [51]

    Guidelines: • The answer NA means that there is no societal impact of the work performed

    Broader Impacts Question: Does the paper discuss both potential positive societal impacts and negative societal impacts of the work performed? Answer: [NA] Justification: This application paper do not talk about broader impacts. Guidelines: • The answer NA means that there is ...

  44. [52]

    Guidelines: • The answer NA means that the paper poses no such risks

    Safeguards Question: Does the paper describe safeguards that have been put in place for responsible release of data or models that have a high risk for misuse (e.g., pretrained language models, image generators, or scraped datasets)? Answer: [NA] Justification: There is no soc...

  45. [53]

    Guidelines: • The answer NA means that the paper does not use existing assets

    Licenses for existing assets Question: Are the creators or original owners of assets (e.g., code, data, models), used in the paper, properly credited and are the license and terms of use explicitly mentioned and properly respected? Answer: [Yes] Justification: The creators and...

  46. [54]

    Guidelines: • The answer NA means that the paper does not release new assets

    New Assets Question: Are new assets introduced in the paper well documented and is the documentation provided alongside the assets? Answer: [Yes] Justification: We give our code, model, training detail and licence in the supplementary material and would provide it in a structu...

  47. [55]

    Crowdsourcing and Research with Human Subjects Question: For crowdsourcing experiments and research with human subjects, does the paper include the full text of instructions given to participants and screenshots, if applicable, as well as details about compensation (if any)? A...

  48. [56]

    Guidelines: • The answer NA means that the paper does not involve crowdsourcing nor research with human subjects

    Institutional Review Board (IRB) Approvals or Equivalent for Research with Human Subjects Question: Does the paper describe potential risks incurred by study participants, whether such risks were disclosed to the subjects, and whether Institutional Review Board (IRB) approvals...

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.