LM-MCVT: A Lightweight Multi-modal Multi-view Convolutional-Vision Transformer Approach for 3D Object Recognition

Hamidreza Kasaei; Songsong Xiong

REVIEW 4 major objections 5 minor 35 references

LM-MCVT: A Lightweight Multi-modal Multi-view Convolutional-Vision Transformer Approach for 3D Object Recognition

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read A lightweight network using four RGB-D views can match or beat view-based 3D classifiers that use twelve or more views, the paper reports on ModelNet and OmniObject3D.

desk verdict Useful lightweight RGBD multi-view classifier with careful ablations, but the state-of-the-art claim is undercut by inconsistent view counts and an undocumented baseline-retraining protocol. read the letter →

arxiv 2504.19256 v3 pith:6KXDYMP4 submitted 2025-04-27 cs.CV cs.RO

Songsong Xiong , Hamidreza Kasaei This is my paper · ORCID

classification cs.CVcs.RO

keywords 3Dobjectrecognitionmulti-viewlearningRGB-Dfusionvisiontransformerentropy-basedlightweightmodelroboticperceptionNet40

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper claims that a lightweight network can recognize 3D objects from only four RGB-D views as accurately as heavier view-based models that use twelve or more views. The proposed LM-MCVT stacks convolutional encoders before and between transformer stages so that local details survive global attention, and it introduces Globally Entropy-based Embeddings Fusion (GEEF), which weights each view by the entropy of its class token before merging RGB and depth features. Across ModelNet10, ModelNet40, and the real-scanned OmniObject3D dataset, the reported accuracies are 98.5%, 95.6%, and 85.1%, with 10.5 million parameters and about 7 milliseconds per RGB-D instance. A dual-arm robot using the fine-tuned model succeeded in ten consecutive pick-and-place trials, which the paper offers as evidence of practical deployability.

What carries the argument

The load-bearing mechanism is GEEF (Globally Entropy-based Embeddings Fusion), which turns each view's class-token entropy into a normalized weight and fuses class tokens by that weight while average-pooling the remaining patch embeddings; RGB and depth streams are fused separately and concatenated. The rest of the architecture exists to feed GEEF good tokens: pre-residual convolutional encoders and middle-residual convolutional encoders restore local detail, local transformers encode patch relations within a view, and global transformers let views exchange information through class tokens. Together they give the model a lightweight profile of 10.5 million parameters.

What would settle it

Retrain MVT and SMV on the same four-view RGB-D renderings and the same ModelNet and OmniObject3D splits, with the same hyperparameter budget, and measure accuracy; if either baseline reaches or exceeds 95.6% on ModelNet40 or 85.1% on OmniObject3D under that protocol, the claimed state-of-the-art result would not hold.

Watch

Extended reading notes

Core claim

The central claim is that coordinated use of convolutional feature extraction, transformer-based global reasoning, and entropy-weighted view fusion produces a 3D object recognizer that is both more accurate and cheaper than the published comparison methods. With four rendered RGB-D views, LM-MCVT reaches 98.5% on ModelNet10, 95.6% on ModelNet40, and 85.1% on OmniObject3D in five-fold cross-validation, while using 10.5 million parameters and 7.0 ms per instance. The paper attributes the improvement over average-pooling fusion to GEEF, which normalizes the entropy of each view's class token into a weight and uses it to combine the views, so views carrying sharper or more distinct information dominate the fused representation.

Load-bearing premise

The accuracy comparison rests on the assumption that MVT and SMV were retrained under exactly the same four-view rendering, the same five-fold splits, and a comparable tuning budget; the paper does not describe that protocol.

Editorial extensions

If this is right

Four views are close to the practical ceiling: on ModelNet10, RGB-D accuracy moves from 98.5% at four views to 98.9% at twelve views while per-instance time grows from 7.0 ms to 29.7 ms.
Depth consistently helps: on ModelNet10, adding depth raises accuracy at every view count, from 93.2% to 95.5% in the single-view case.
The entropy-weighted fusion beats the alternatives it was tested against: GEEF reaches 96.6% on ModelNet10 RGB-D versus 95.4% for the best averaging-based embedding fusion.
The same four-view model transfers to real-scanned data, where it reports 85.1% accuracy under five-fold cross-validation, and a fine-tuned version reached 99.3% on synthetic household objects and succeeded in ten real pick-and-place trials.
At 10.5 million parameters and 7.0 ms per RGB-D instance, the model sits in a practical range for real-time robotic perception.

Reading between the lines

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

If the four-view protocol is taken at face value, the implication extends beyond classification: dense multi-view capture may be unnecessary whenever depth is available, so latency budgets in robotics can be spent on other perception stages.
GEEF is a generic fusion operator: it could be applied wherever multiple views or frames produce class tokens, such as video action recognition, multi-camera detection, or active perception, where viewpoint informativeness varies.
A natural test of the paper's mechanistic claim is to visualize the learned entropy weights per view; if high-entropy views are the ambiguous ones, entropy weighting is doing uncertainty-aware pooling, whereas if high-entropy views are the informative ones, it is doing saliency weighting.
The absence of a described baseline protocol means the strongest testable extension is an independent reproduction with matched views, splits, and tuning, which would separate the contribution of the architecture from the contribution of the evaluation setup.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: this is a competent engineering paper with a nice robot demo, but the headline claim that LM-MCVT "sets a new benchmark" is not supported by the evidence as presented. The architecture is a mix of known ingredients — residual convs around a ViT, local/global transformer pairs, class-token fusion — and the main novelty, GEEF, is entropy-weighted fusion over class tokens, which is a standard idea with an underspecified formula. What the paper does well: the ablation study is thorough (fusion variants, component removal, view-count and view-structure robustness), the four-view RGBD setting is practically relevant, and the real-robot pick-and-place test is a genuine attempt at validation. The efficiency numbers (10.5M params, 7 ms) are attractive.

The soft spots are real and load-bearing for the SOTA claim. First, Eq. (1) defines H as −Σ p log p but never says what p is. The class token is an embedding, not a distribution; until p is defined the equation is incomplete. Second, the paper says comparisons were made under "consistent viewpoint configurations," but Table VI lists all published baselines at 12 views while LM-MCVT uses 4. If those numbers are copied from the original papers, that is fine as a reference table, but it is not a same-protocol comparison. Third, on OmniObject3D the only protocol statement is "we trained all multi-view methods from scratch" — no view count, no splits, no hyperparameters, no mention of official code. Since MVT and SMV were designed around 12 views, feeding them 4 could handicap them. The 4–15 point margins on OmniObject3D are suspicious precisely because the protocol is invisible. Fourth, the ModelNet40 margin over VGP is 0.29 points, which is within run-to-run noise; no standard deviation is reported even for the 5-fold OmniObject3D result.

None of this makes the method useless. If the authors disclose the full retraining setup and release code or checkpoints, the empirical claims can be checked. As it stands, the paper is a solid engineering contribution with unreliable comparison claims. I would send it to peer review but with a demand for a complete protocol description, same-view-count comparisons (or an explicit statement that baselines use published 12-view numbers), a definition of p in GEEF, and variance estimates. The robot demo is a plus but does not substitute for a sound comparison.

Who this is for: people building lightweight RGBD multi-view classifiers for robotics. They will get value from the architecture and ablations even if the SOTA claim is discounted. I would not cite the headline numbers until the protocol issue is resolved, but I would cite the architecture/ablation parts in a related work section.

Recommendation: conditional accept with major revisions; the core method is reasonable but the evaluation needs to be made honest and reproducible.

Referee Report

4 major / 5 minor

Summary. The paper proposes LM-MCVT, a lightweight multi-view RGBD 3D object recognition architecture that combines pre- and middle-residual convolutional encoders with local and global transformers, and introduces a Globally Entropy-based Embeddings Fusion (GEEF) method for view aggregation. Experiments on ModelNet10, ModelNet40, and OmniObject3D report accuracies of 98.5%, 95.6%, and 85.1% respectively, using only four views and 10.5M parameters, together with a real-robot demonstration. The central claim is that LM-MCVT surpasses state-of-the-art view-based methods in both accuracy and efficiency.

Significance. If fully substantiated, the paper would make a useful practical contribution: a four-view RGBD model with 10.5M parameters and 7.0 ms inference time that outperforms much larger view-based models such as MVT and SMV on OmniObject3D, and that matches or beats 12-view baselines on ModelNet40. The component ablations and the viewpoint-structure robustness experiments are informative, and the real-robot deployment adds credibility to the efficiency claims. However, the significance currently rests on three unverified pillars: the GEEF entropy formula must be well-defined and reproducible, the OmniObject3D baseline comparisons must be shown to be fair, and the 5-fold results must be accompanied by variance information. The paper also deserves credit for reporting parameter counts and inference times, and for including ablations of the fusion strategy and architectural components.

major comments (4)

[Section III-E, Eq. (1)] The entropy H^cls_v(c_vj) in Eq. (1) is not computable as written. No probability distribution p is defined; the summation index j runs over views, while the argument c_vj is a single class-token vector, so the notation mixes per-view and global quantities. Since GEEF is the paper's central contribution and the weights w_vj are used in Eq. (2), the authors must define p explicitly (for example, as a softmax over the class-token vector components or over the view dimension) and correct the index structure. Without this, the fusion method cannot be reproduced or compared against alternatives.
[Section IV-E-2, Table VII] The protocol for retraining MVT and SMV on OmniObject3D is insufficient. The statement 'we trained all multi-view methods from scratch' leaves unspecified the number of views presented to each baseline, the exact 5-fold split assignment, whether official implementations or re-implementations were used, and the hyperparameter search budget. Since MVT and SMV were originally designed for 12-view inputs, feeding them four views may systematically handicap their view-selection mechanisms. The large OmniObject3D margins (85.1 vs. 70.7 for MVT and 81.0 for SMV) therefore cannot be taken as evidence of superiority until the comparison is shown to be fair. Please report the full training and evaluation protocol, or temper the state-of-the-art claim accordingly.
[Section IV-A and Table VII] The 5-fold cross-validation on OmniObject3D is reported only as single accuracy numbers, with no standard deviation, per-fold breakdown, or number of random seeds. Without this information, the reader cannot assess whether the difference between LM-MCVT (RGBD, 85.1%) and SMV (RGBD, 81.0%) is stable or within run-to-run variation. Please add per-fold results and variance estimates, or at least multiple-seed statistics, to support the robustness claim.
[Section IV-E-1, Table VI] The claim of a new benchmark on ModelNet40 is not supported by a same-view-count comparison. All view-based baselines in Table VI use 12 views (MVCNN uses 80), while LM-MCVT uses four, which contradicts the paper's statement in the introduction to Section IV that comparisons were made 'under consistent viewpoint configurations.' Moreover, the 0.29-point margin over VGP on ModelNet40 is within typical run-to-run variance, and no standard deviation is reported. A fair comparison would include baselines evaluated at four views, or at least a discussion of how view count affects each baseline's accuracy.

minor comments (5)

[Table VI] The merged 'Params' and 'Views' cells for the three LM-MCVT rows are easy to misread; please make explicit that Params = 10.5M and Views = 4 apply to all three input modalities.
[Fig. 6 caption] The caption contains the typo 'Rivz'; this should be 'RViz'.
[Section IV-D, Tables IV-V] The five random four-view structures are shown in figures, but the exact elevation angles and in-plane rotations are not specified; please state the generation procedure so that the robustness claim is reproducible.
[Section III-D] The description of how 'synthesized patch features' are formed is ambiguous because the middle-residual block excludes class tokens while the global transformer input later concatenates them; please clarify the tensor shapes and the concatenation order.
[Section IV-B] The fusion-strategy ablation is performed only on ModelNet10 and with the pre- and middle-residual encoders removed; please note this limitation explicitly, since the behavior of GEEF with the full architecture is not separately ablated.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the paper is an empirical architecture study evaluated against external benchmarks, and its self-citations are contextual only.

full rationale

This is an empirical architecture paper, not a derivation. The central claims are accuracies measured on ModelNet10, ModelNet40, and OmniObject3D against published or retrained baselines; no theoretical result is derived from the method's own equations. GEEF is a fusion rule defined in Eqs. (1)-(2) from class-token entropies; it is introduced and then ablated on ModelNet10, so its performance is an experimental finding, not a prediction equivalent to its construction. The state-of-the-art comparisons use external methods such as MVCNN, MVT, SMV, and VGP, and the paper's self-citations ([15], [31], [34]) appear only as contextual related work, as the source of the Synthetic Household Objects dataset for robotic fine-tuning, and as a grasping module in the demonstration; none supplies the recognition accuracy on which the SOTA claim rests. The baseline-retraining protocol and view-count consistency concerns are experimental-fairness/correctness issues, not circularity; therefore the circularity score is 0.

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

The method introduces no fitted scalar parameters; its free choices are architectural (block counts, view count) selected by validation on ModelNet10. The technical contribution GEEF introduces no new physical or mathematical entities, but its operation depends on two unstated premises about entropy of class-token vectors.

free parameters (4)

view count (l) = 4
Selected from ModelNet10 experiments in Section IV-D (Table III) as the accuracy/time tradeoff point, then reused for ModelNet40 and OmniObject3D; the headline numbers are tied to this hand-picked value.
pre-residual encoder count (M) = 2
Chosen from the Fig. 3 ablation as part of the 'optimal balance' configuration in Section IV-C; not independently justified.
middle-residual encoder count (S) = 7
Selected from the same Fig. 3 ablation in Section IV-C as the 'optimal balance' configuration.
transformer depths = 8 local, 4 global
Inherited from MVT [14] and fixed without sensitivity analysis for this specific architecture.

assumptions (4)

ad hoc to paper Class-token entropy is a valid view-quality signal for fusion weighting.
GEEF's premise in Section III-E is that views with higher class-token entropy deserve more weight; this is asserted, not demonstrated, and no prior reference is cited for it.
ad hoc to paper The entropy in Eq. (1) is computable from a well-defined probability distribution over the class-token vector.
The symbol p in H(c_vj) = -sum p log p is never defined in Section III-E; any implementation must invent a distribution, and different choices change the fusion weights and the reported gains.
domain assumption Rendered 2D RGBD views of an object carry enough information to classify it.
Standard premise of view-based methods inherited from MVCNN [27] and MVT [14]; the paper relies on it in Section III-A.
ad hoc to paper Baselines retrained from scratch in Table VII received equivalent views, splits, and tuning effort.
Section IV-E-2 states all multi-view methods were trained from scratch but gives no protocol; the fairness of the comparison is an unverified internal assumption.

how reviews work

0 comments

Cite this review

Pith. "Pith review of LM-MCVT: A Lightweight Multi-modal Multi-view Convolutional-Vision Transformer Approach for 3D Object Recognition." pith.science (2026). https://pith.science/paper/6KXDYMP4

@misc{pith2026250419256,
  author       = {Pith},
  title        = {Pith review of: LM-MCVT: A Lightweight Multi-modal Multi-view Convolutional-Vision Transformer Approach for 3D Object Recognition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6KXDYMP4}},
  note         = {Machine review of arXiv:2504.19256}
}

read the original abstract

In human-centered environments such as restaurants, homes, and warehouses, robots often face challenges in accurately recognizing 3D objects. These challenges stem from the complexity and variability of these environments, including diverse object shapes. In this paper, we propose a novel Lightweight Multi-modal Multi-view Convolutional-Vision Transformer network (LM-MCVT) to enhance 3D object recognition in robotic applications. Our approach leverages the Globally Entropy-based Embeddings Fusion (GEEF) method to integrate multi-views efficiently. The LM-MCVT architecture incorporates pre- and mid-level convolutional encoders and local and global transformers to enhance feature extraction and recognition accuracy. We evaluate our method on the synthetic ModelNet40 dataset and achieve a recognition accuracy of 95.6% using a four-view setup, surpassing existing state-of-the-art methods. To further validate its effectiveness, we conduct 5-fold cross-validation on the real-world OmniObject3D dataset using the same configuration. Results consistently show superior performance, demonstrating the method's robustness in 3D object recognition across synthetic and real-world 3D data.

Figures

Figures reproduced from arXiv: 2504.19256 by the authors.

**Figure 2.** This section provides a detailed explanation of each [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 2.** LM-MCVT Framework for 3D Object Recognition in Robotic Perception. The 3D object, i.e., [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Figures from the paper (4 more)

**Figure 3.** Figure 3: Ablation study on the effects of pre-residual and middle-residual [PITH_FULL_IMAGE:figures/full_fig_p005_3.png]

**Figure 4.** Figure 4: Visualization of five cases of four-view structures with circular [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Visualization of five cases of four-view structures with hemi [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

**Figure 6.** Figure 6: The snapshots demonstrating the recognition performance of our dual-arm robot employing the LM-MCVT model in [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 25 canonical work pages

[1]

Lightweight object detection algorithm for robots with improved yolov5,

G. Liu, Y . Hu, Z. Chen, J. Guo, and P. Ni, “Lightweight object detection algorithm for robots with improved yolov5,” Engineering Applications of Artificial Intelligence , vol. 123, p. 106217, 2023

work page 2023
[2]

V oxnet: A 3d convolutional neural net- work for real-time object recognition,

D. Maturana and S. Scherer, “V oxnet: A 3d convolutional neural net- work for real-time object recognition,” in 2015 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, 2015, pp. 922–928

work page 2015
[3]

Fuzzy lightweight cnn for point cloud object clas- sification based on voxel,

O. V . Putra, M. I. Riansyah, A. Priyadi, E. M. Yuniarno, M. H. Purnomo et al. , “Fuzzy lightweight cnn for point cloud object clas- sification based on voxel,” in TENCON 2023-2023 IEEE Region 10 Conference (TENCON). IEEE, 2023, pp. 685–690

work page 2023
[4]

Similarity measurement and retrieval of three-dimensional voxel model based on symbolic operator,

Z. He, X. Liu, and C. Zhang, “Similarity measurement and retrieval of three-dimensional voxel model based on symbolic operator,” ISPRS International Journal of Geo-Information , vol. 13, no. 3, p. 89, 2024

work page 2024
[5]

Pointnet: Deep learning on point sets for 3d classification and segmentation,

C. R. Qi, H. Su, K. Mo, and L. J. Guibas, “Pointnet: Deep learning on point sets for 3d classification and segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2017, pp. 652–660

2017
[6]

Pointramba: A hybrid transformer-mamba framework for point cloud analysis,

Z. Wang, Z. Chen, Y . Wu, Z. Zhao, L. Zhou, and D. Xu, “Pointramba: A hybrid transformer-mamba framework for point cloud analysis,” arXiv preprint arXiv:2405.15463 , 2024

arXiv 2024
[7]

Masked autoencoders for point cloud self-supervised learning,

Y . Pang, W. Wang, F. E. Tay, W. Liu, Y . Tian, and L. Yuan, “Masked autoencoders for point cloud self-supervised learning,” in European conference on computer vision . Springer, 2022, pp. 604–621

2022
[8]

Point could mamba: Point cloud learning via state space model,

T. Zhang, X. Li, H. Yuan, S. Ji, and S. Yan, “Point could mamba: Point cloud learning via state space model,” arXiv preprint arXiv:2403.00762, 2024

arXiv 2024

Show all 35 references

[9]

Multi-view saliency guided deep neural network for 3-d object retrieval and classification,

H.-Y . Zhou, A.-A. Liu, W.-Z. Nie, and J. Nie, “Multi-view saliency guided deep neural network for 3-d object retrieval and classification,” IEEE Transactions on Multimedia , vol. 22, no. 6, pp. 1496–1506, 2019

2019
[10]

Review of multi-view 3d object recognition methods based on deep learning,

S. Qi, X. Ning, G. Yang, L. Zhang, P. Long, W. Cai, and W. Li, “Review of multi-view 3d object recognition methods based on deep learning,” Displays, vol. 69, p. 102053, 2021

2021
[11]

Simultaneous multi-view object recognition and grasping in open-ended domains,

H. Kasaei, M. Kasaei, G. Tziafas, S. Luo, and R. Sasso, “Simultaneous multi-view object recognition and grasping in open-ended domains,” Journal of Intelligent & Robotic Systems , vol. 110, no. 2, pp. 1–19, 2024

2024
[12]

Selective multi- view deep model for 3d object classification,

M. Alzahrani, M. Usman, S. Anwar, and T. Helmy, “Selective multi- view deep model for 3d object classification,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, June 2024, pp. 728–736

2024
[13]

Ovpt: Optimal viewset pooling transformer for 3d object recognition,

W. Wang, G. Chen, H. Zhou, and X. Wang, “Ovpt: Optimal viewset pooling transformer for 3d object recognition,” in Proceedings of the Asian Conference on Computer Vision , 2022, pp. 4444–4461

2022
[14]

MVT: multi-view vision transformer for 3d object recognition,

S. Chen, T. Yu, and P. Li, “MVT: multi-view vision transformer for 3d object recognition,” in BMVC, 2021

2021
[15]

Early or late fusion matters: Efficient rgb- d fusion in vision transformers for 3d object recognition,

G. Tziafas and H. Kasaei, “Early or late fusion matters: Efficient rgb- d fusion in vision transformers for 3d object recognition,” in 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2023, pp. 9558–9565

2023
[16]

Multi-view attention-convolution pooling network for 3d point cloud classification,

W. Wang, T. Wang, and Y . Cai, “Multi-view attention-convolution pooling network for 3d point cloud classification,” Applied Intelli- gence, vol. 52, no. 13, pp. 14 787–14 798, 2022

2022
[17]

Learning to select views for efficient multi-view understanding,

Y . Hou, S. Gould, and L. Zheng, “Learning to select views for efficient multi-view understanding,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024, pp. 20 135–20 144

2024
[18]

Do vision transformers see like convolutional neural networks?

M. Raghu, T. Unterthiner, S. Kornblith, C. Zhang, and A. Dosovitskiy, “Do vision transformers see like convolutional neural networks?” Advances in Neural Information Processing Systems , vol. 34, pp. 12 116–12 128, 2021

2021
[19]

Antipodal robotic grasping using generative residual convolutional neural network,

S. Kumra, S. Joshi, and F. Sahin, “Antipodal robotic grasping using generative residual convolutional neural network,” in 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2020, pp. 9626–9633

2020
[20]

Mvsalnet: Multi-view augmentation for rgb-d salient object detection,

J. Zhou, L. Wang, H. Lu, K. Huang, X. Shi, and B. Liu, “Mvsalnet: Multi-view augmentation for rgb-d salient object detection,” in Euro- pean Conference on Computer Vision . Springer, 2022, pp. 270–287

2022
[21]

Dominant set clustering and pooling for multi-view 3d object recognition,

C. Wang, M. Pelillo, and K. Siddiqi, “Dominant set clustering and pooling for multi-view 3d object recognition,” arXiv preprint arXiv:1906.01592, 2019

1906 arXiv
[22]

3d shapenets: A deep representation for volumetric shapes,

Z. Wu, S. Song, A. Khosla, F. Yu, L. Zhang, X. Tang, and J. Xiao, “3d shapenets: A deep representation for volumetric shapes,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, pp. 1912–1920

2015
[23]

Omniobject3d: Large-vocabulary 3d object dataset for realistic perception, reconstruction and generation,

T. Wu, J. Zhang, X. Fu, Y . Wang, J. Ren, L. Pan, W. Wu, L. Yang, J. Wang, C. Qian et al. , “Omniobject3d: Large-vocabulary 3d object dataset for realistic perception, reconstruction and generation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Reco...

2023
[24]

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

A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly et al., “An image is worth 16x16 words: Transformers for image recognition at scale,” arXiv preprint arXiv:2010.11929 , 2020

2010 arXiv
[25]

Point convolutional neural networks by extension operators,

M. Atzmon, H. Maron, and Y . Lipman, “Point convolutional neural networks by extension operators,” arXiv preprint arXiv:1803.10091 , 2018

2018 arXiv
[26]

A voxelized point clouds representation for object classification and segmentation on 3d data,

A. S. Gezawa, Z. A. Bello, Q. Wang, and L. Yunqi, “A voxelized point clouds representation for object classification and segmentation on 3d data,” The Journal of Supercomputing , vol. 78, no. 1, pp. 1479–1500, 2022

2022
[27]

Multi- view convolutional neural networks for 3d shape recognition,

H. Su, S. Maji, E. Kalogerakis, and E. Learned-Miller, “Multi- view convolutional neural networks for 3d shape recognition,” in Proceedings of the IEEE international conference on computer vision , 2015, pp. 945–953

2015
[28]

Mvtn: Learn- ing multi-view transformations for 3d understanding,

A. Hamdi, F. AlZahrani, S. Giancola, and B. Ghanem, “Mvtn: Learn- ing multi-view transformations for 3d understanding,” International Journal of Computer Vision , pp. 1–30, 2024

2024
[29]

Tensor multi-task learning for multi- view representation of 3d shape,

Y . Zhang, S. Fu, and Y . Qu, “Tensor multi-task learning for multi- view representation of 3d shape,” in 2023 China Automation Congress (CAC). IEEE, 2023, pp. 4703–4708

2023
[30]

A viewpoint- guided prototype network for 3d shape classification,

L. Han, J. He, F. Dou, H. Ma, X. Xie, and W. Yang, “A viewpoint- guided prototype network for 3d shape classification,” Multimedia Systems, vol. 29, no. 6, pp. 3531–3547, 2023

2023
[31]

Lifelong ensemble learning based on multiple representations for few-shot object recognition,

H. Kasaei and S. Xiong, “Lifelong ensemble learning based on multiple representations for few-shot object recognition,” Robotics and Autonomous Systems, vol. 174, p. 104615, 2024

2024
[32]

Towards lifelong assistive robotics: A tight coupling between object perception and manipulation,

S. H. Kasaei, M. Oliveira, G. H. Lim, L. S. Lopes, and A. M. Tom ´e, “Towards lifelong assistive robotics: A tight coupling between object perception and manipulation,” Neurocomputing, vol. 291, pp. 151–166, 2018

2018
[33]

Orthographicnet: A deep transfer learning approach for 3-d object recognition in open-ended domains,

S. H. Kasaei, “Orthographicnet: A deep transfer learning approach for 3-d object recognition in open-ended domains,” IEEE/ASME Transactions on Mechatronics, vol. 26, no. 6, pp. 2910–2921, 2020

2020
[34]

Mvgrasp: Real-time multi-view 3d object grasping in highly cluttered environments,

H. Kasaei and M. Kasaei, “Mvgrasp: Real-time multi-view 3d object grasping in highly cluttered environments,” Robotics and Autonomous Systems, vol. 160, p. 104313, 2023

2023
[35]

Towards open-world grasping with large vision-language models,

G. Tziafas and H. Kasaei, “Towards open-world grasping with large vision-language models,” in 8th Annual Conference on Robot Learn- ing, 2024

2024

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [1]

Lightweight object detection algorithm for robots with improved yolov5,

G. Liu, Y . Hu, Z. Chen, J. Guo, and P. Ni, “Lightweight object detection algorithm for robots with improved yolov5,” Engineering Applications of Artificial Intelligence , vol. 123, p. 106217, 2023

work page 2023

[2] [2]

V oxnet: A 3d convolutional neural net- work for real-time object recognition,

D. Maturana and S. Scherer, “V oxnet: A 3d convolutional neural net- work for real-time object recognition,” in 2015 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, 2015, pp. 922–928

work page 2015

[3] [3]

Fuzzy lightweight cnn for point cloud object clas- sification based on voxel,

O. V . Putra, M. I. Riansyah, A. Priyadi, E. M. Yuniarno, M. H. Purnomo et al. , “Fuzzy lightweight cnn for point cloud object clas- sification based on voxel,” in TENCON 2023-2023 IEEE Region 10 Conference (TENCON). IEEE, 2023, pp. 685–690

work page 2023

[4] [4]

Similarity measurement and retrieval of three-dimensional voxel model based on symbolic operator,

Z. He, X. Liu, and C. Zhang, “Similarity measurement and retrieval of three-dimensional voxel model based on symbolic operator,” ISPRS International Journal of Geo-Information , vol. 13, no. 3, p. 89, 2024

work page 2024

[5] [5]

Pointnet: Deep learning on point sets for 3d classification and segmentation,

C. R. Qi, H. Su, K. Mo, and L. J. Guibas, “Pointnet: Deep learning on point sets for 3d classification and segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2017, pp. 652–660

2017

[6] [6]

Pointramba: A hybrid transformer-mamba framework for point cloud analysis,

Z. Wang, Z. Chen, Y . Wu, Z. Zhao, L. Zhou, and D. Xu, “Pointramba: A hybrid transformer-mamba framework for point cloud analysis,” arXiv preprint arXiv:2405.15463 , 2024

arXiv 2024

[7] [7]

Masked autoencoders for point cloud self-supervised learning,

Y . Pang, W. Wang, F. E. Tay, W. Liu, Y . Tian, and L. Yuan, “Masked autoencoders for point cloud self-supervised learning,” in European conference on computer vision . Springer, 2022, pp. 604–621

2022

[8] [8]

Point could mamba: Point cloud learning via state space model,

T. Zhang, X. Li, H. Yuan, S. Ji, and S. Yan, “Point could mamba: Point cloud learning via state space model,” arXiv preprint arXiv:2403.00762, 2024

arXiv 2024