Pith. sign in

REVIEW 4 major objections 5 minor 47 references

You Only Estimate Once: Unified, One-stage, Real-Time Category-level Articulated Object 6D Pose Estimation for Robotic Grasping

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A single 200 Hz pass estimates articulated-part 6D poses.

desk verdict A genuinely new single-stage architecture for articulated part pose with a strong joint-training ablation, but the headline accuracy claim overstates a worse translation error and the real-time claim skips the upstream detector. read the letter →

arxiv 2506.05719 v1 pith:WH27OOY7 submitted 2025-06-06 cs.CV cs.RO

Jingshun Huang , Haitao Lin , Tianyu Wang , Yanwei Fu , Yu-Gang Jiang , Xiangyang Xue This is my paper
classification cs.CVcs.RO
keywords 6Dposeestimationarticulatedobjectscategory-levelroboticgraspingpointcloudinstancesegmentationNPCSsingle-stagenetworkreal-timemanipulation
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

This paper claims that category-level pose estimation for articulated objects does not need a two-stage pipeline that first segments part instances and then estimates Normalized Part Coordinate Space (NPCS) representations. YOEO is a single-stage point-cloud network that jointly predicts per-point semantic labels, offsets to part-instance centroids, and NPCS coordinates, then clusters points by their voted centroids and aligns each NPCS region to the observed cloud with Umeyama registration to recover pose and size. On the GAPart dataset, the authors report improved rotation error (9.0 degrees versus 9.9 degrees) and 3D mean intersection-over-union (57.6 percent versus 51.2 percent) over the two-stage GAPartNet baseline, with only 1.9 million parameters and inference at 200 Hz. They also deploy the synthetically trained model on a real robot to manipulate unseen articulated objects, which demonstrates that a single forward pass can support real-time robotic perception.

What carries the argument

The load-bearing object is the unified one-stage architecture: a RandLA-Net encoder-decoder whose features feed three parallel shared-MLP heads—semantic labels trained with focal loss, per-point centroid offsets trained with L1 loss, and NPCS coordinates trained as 100-bin classification per axis. The mechanism that separates instances is centroid voting: every point predicts a vector to its part's centroid, so points of the same instance agree and different instances of the same semantic class separate in offset space; a clustering step groups them, and RANSAC plus Umeyama least-squares registration recovers the similarity transform from the predicted NPCS region to the observed point cloud.

What would settle it

Run YOEO on a fixed set of GAPart test objects using two input sources: object crops from Grounding-DINO text prompts and ground-truth object masks from the dataset. If rotation error and mIoU stay comparable across the two, the upstream dependency is weak; if accuracy drops sharply for small or metallic instances or under occlusion, the dependency is confirmed. A complementary test is to artificially decimate or partially mask the input point cloud and record at what coverage the rotation error exceeds 15 degrees.

Watch

Extended reading notes

Core claim

The central claim is that a single forward pass of one unified network can replace the cascade of part segmentation followed by per-part NPCS estimation, and that co-training the three prediction heads—semantic labeling, centroid-offset voting, and NPCS regression—is what produces the accuracy gain rather than a more complex architecture. The paper's evidence is the ablation study: parallel training reduces rotation error from 19.6 degrees to 9.0 degrees, reduces translation and scale errors from 0.14 cm to 0.11 cm and from 0.041 cm to 0.036 cm, and lifts 3D mIoU from 52.3 percent to 57.6 percent compared to individually trained heads. The full model reaches 200 Hz inference with 1.9 million parameters, and a physical Kinova robot completes 22 of 30 manipulation trials on objects not seen during training.

Load-bearing premise

The load-bearing premise is that the point cloud delivered to the network is a clean, complete crop of the target object from Grounding-DINO, and that the depth sensor provides usable geometry for the part; failures of that upstream detector or depth acquisition are not evaluated, and the paper's own limitations concede poor performance on small and metallic objects.

Editorial extensions

If this is right

  • A one-stage formulation removes the error accumulation that comes from cascaded segmentation-then-pose pipelines.
  • Co-training three auxiliary heads is a practical recipe for improving NPCS regression accuracy without adding new supervision.
  • Because the predicted NPCS frame carries the joint or prismatic axis and predefined grasp poses, the same registered transform directly yields robot motion commands.
  • At 200 Hz and 1.9 million parameters, the perception cost drops enough for on-board, real-time control loops rather than offline pose readouts.

Reading between the lines

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

  • The large jump between individual and parallel rotation error suggests the auxiliary heads act as a learned regularizer for the NPCS head; one could test this by adding the auxiliary losses incrementally to a pure NPCS network and measuring the gain at each step.
  • The pipeline's reliance on Grounding-DINO for the input crop means the whole system's accuracy is upper-bounded by the detector's recall; an evaluation that reports pose error conditioned on detector success versus failure would expose how much of the sim-to-real result depends on that upstream stage.
  • Since the paper lists small and metallic objects as failure cases due to depth quality, fusing RGB cues from the same crop is a natural extension, and its value could be quantified by measuring rotation error on those object classes with and without RGB fusion.
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 / 5 minor

Summary. The paper proposes YOEO, a single-stage category-level articulated-object 6D pose estimation method. A RandLA-Net backbone jointly predicts per-point semantic labels, instance centroid offsets, and Normalized Part Coordinate Space (NPCS) maps; a clustering step groups points into part instances, and Umeyama SIM(3) registration between predicted NPCS points and observed points recovers pose and size. Experiments on the GAPartNet dataset compare against PointGroup- and AutoGPart-based baselines and GAPartNet, and a real-robot study on a Kinova arm reports manipulation success rates. The paper claims faster inference (200 Hz), fewer parameters (1.9 M), and improved pose accuracy, particularly in mIoU.

Significance. If the claims are substantiated, YOEO would be a useful step toward practical real-time perception for articulated-object manipulation: a single lightweight network that simultaneously outputs segmentation and NPCS maps is architecturally simpler than two-stage pipelines, and the ablation in Table II provides concrete evidence that joint training of the three heads improves pose accuracy relative to individually trained heads. The real-robot deployment is also a valuable proof-of-concept. However, the central accuracy claim is currently overstated relative to the paper's own Table I, and the end-to-end real-time claim depends on an unvalidated Grounding-DINO upstream component. The core idea is defensible, but several load-bearing claims need revision or additional experiments.

major comments (4)
  1. [Table I and Section IV-A] The claim that YOEO achieves 'significantly improved pose accuracy compared to the previous state-of-the-art method, GAPartNet' is contradicted by the translation error in Table I: Te is 0.11 cm for YOEO versus 0.024 cm for GAPartNet, a 4.6x degradation, while rotation error improves only marginally (9.0 vs 9.9 degrees) and A5/A10 improve by only 2.1 and 1.3 percentage points. The large mIoU gain (57.6 vs 51.2) is a segmentation metric, not a pose accuracy metric. The authors should either explain why the translation error is worse (e.g., a quirk of the normalized scale or of the clustering step) or substantially qualify the headline accuracy claim.
  2. [Section III (Architecture Overview) and Section IV (Evaluation)] The input point cloud is stated to be 'obtained from the output of the segmentation model. Here, we use Grounding-DINO,' but no experiment measures the impact of this upstream detector. If Grounding-DINO misses the object, returns a partial crop, or receives an unmatched text prompt, the pose estimate is undefined or degraded; the paper's own Limitations section concedes poor behavior on small and metallic objects, which is precisely the regime where detection and depth are unreliable. The reported 200 Hz speed is 'calculated from feeding object point clouds to get part poses' (Section IV), so it excludes Grounding-DINO, clustering, RANSAC, and camera/robot overhead and is not an end-to-end rate. The authors should either evaluate the full pipeline including Grounding-DINO or clearly restate the claimed speed as network-only.
  3. [Section IV-C, Table III] The real-robot results are too weak to support the utility claim as stated. YOEO achieves 22/30 total successes versus 20/30 for GAPartNet, but on the drawer task YOEO is worse (5/10 vs 7/10). With 10 trials per task, these differences are not statistically meaningful, and the paper does not report a breakdown of failures into upstream detection failures, pose estimation failures, or execution failures. Without such a breakdown, the 22/30 rate cannot be attributed specifically to YOEO's pose estimation. The authors should add a failure analysis and, ideally, more trials.
  4. [Section IV-B, Table II] The ablation description is underspecified. The paper says each head is 'trained individually by freezing the other two' and then the individually trained heads are combined, but it does not state which backbone/feature-extraction parameters are frozen, whether each individually trained head uses the same trained backbone, or how the combined model is assembled at inference. Since the joint-training benefit is a central contribution, this procedure must be described precisely enough to be reproducible. The authors should clarify the training protocol and, if possible, report the variance across multiple training runs.
minor comments (5)
  1. [Throughout] The baseline name is inconsistently spelled 'GAPartNet' and 'GAParNet' in the abstract, introduction, and Table I captions; please use 'GAPartNet' consistently.
  2. [Section III-A, Eq. (1)] The focal loss definition is unclear: 'qi = ci · li' mixes confidence with a one-hot label; please define qi explicitly as the predicted probability for the ground-truth class, or write the loss in the standard form.
  3. [Section III-C] The text cites reference [44] for both RANSAC and Umeyama, but reference [44] is the RANSAC paper while Umeyama is reference [16]; the citations should be corrected.
  4. [Table I and Table II] The metric notation 'A5=5◦5cm' and 'A10=10◦10cm' is missing multiplication signs and spaces; please format as '5°, 5 cm' for readability.
  5. [Section IV] The paper states that object scales in the dataset are normalized to a range of 0 to 1 cm; this unusual normalization should be explained, since it directly affects the interpretation of Te and Se values.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the pose and size estimates are produced by a standard least-squares registration of network-predicted NPCS maps against observed points, trained from dataset ground truth rather than from the fitted outputs.

full rationale

The paper's derivation chain is self-contained and does not reduce to its own inputs. The NPCS representation is adopted from prior published work [11, 14] and is predicted by a network trained with softmax cross-entropy against GAPartNet ground-truth canonical coordinates. The final 6D pose and size are recovered by Umeyama SIM(3) registration between the predicted NPCS point set and the observed point cloud, which is a standard least-squares procedure and is not constructed from the same parameters being reported as predictions. No loss term or evaluation metric is defined in terms of the fitted pose or size outputs, so there is no self-definitional or fitted-input-called-prediction step. The ablation study comparing individual and parallel training is an empirical test of the joint-training hypothesis, not a tautology, and its results (Re 19.6 to 9.0) are presented as experimental evidence rather than as a consequence of the method's definitions. Self-citations appear only in the related-work discussion (e.g., SAR-Net, sketch-based grasping, pouring) and are not used to justify the central pose-estimation claim, so they are not load-bearing. The paper does openly state limitations about small and metallic objects, and the reported 200 Hz speed is scoped as being computed from object point clouds rather than including the upstream Grounding-DINO segmentation; these are evaluation-coverage and completeness concerns, not circularity. Therefore no circular step can be quoted or exhibited, and the appropriate score is 0.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The method is supervised learning; its free parameters are design choices (NPCS bin count, focal loss alpha/gamma, clustering threshold) that affect accuracy but are not reported in the paper. The axioms are the upstream Grounding-DINO dependency, the learnability of category-level NPCS, and the standard Umeyama registration. No new physical entities (forces, particles, dimensions) are introduced.

free parameters (3)
  • NPCS coordinate discretization bins = 100 bins per axis
    Section III-C discretizes predicted NPCS coordinates into 100 bins per axis; the bin count is a hand-chosen hyperparameter controlling precision of the canonical coordinate predictions.
  • Focal loss alpha and gamma = not reported
    Equation (1) uses alpha and gamma as balancing and focusing parameters, but their values are not given in the paper.
  • Centroid clustering distance threshold = not reported
    The clustering algorithm is not named and no distance threshold or linkage criterion is reported; this hyperparameter directly affects instance segmentation quality and is chosen by hand.
assumptions (3)
  • domain assumption The input to the network is a point cloud of a single object produced by Grounding-DINO segmentation, not the full scene.
    Section III Architecture Overview states 'we use Grounding-DINO ... to generate the input point cloud'; if the detector fails to isolate the object, the pose estimate is undefined or degraded, and this dependency is not evaluated.
  • domain assumption A category-level NPCS representation exists and is learnable across object categories, mapping all instances of a part category into a canonical frame.
    Adopted from GAPartNet [14] and ANCSH [11]; the paper assumes the normalized canonical space generalizes to unseen instances, which is the core premise of the method.
  • standard math The Umeyama least-squares registration provides a valid SIM(3) solution from the predicted NPCS points to the observed points when correspondences are known.
    Used in Section III-C to recover {s, R, t}; a standard result, but requires correspondences and a reasonable overlap between the predicted NPCS and observed point clouds.

how reviews work

0 comments
Cite this review

Pith. "Pith review of You Only Estimate Once: Unified, One-stage, Real-Time Category-level Articulated Object 6D Pose Estimation for Robotic Grasping." pith.science (2026). https://pith.science/paper/WH27OOY7

@misc{pith2026250605719,
  author       = {Pith},
  title        = {Pith review of: You Only Estimate Once: Unified, One-stage, Real-Time Category-level Articulated Object 6D Pose Estimation for Robotic Grasping},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WH27OOY7}},
  note         = {Machine review of arXiv:2506.05719}
}
read the original abstract

This paper addresses the problem of category-level pose estimation for articulated objects in robotic manipulation tasks. Recent works have shown promising results in estimating part pose and size at the category level. However, these approaches primarily follow a complex multi-stage pipeline that first segments part instances in the point cloud and then estimates the Normalized Part Coordinate Space (NPCS) representation for 6D poses. These approaches suffer from high computational costs and low performance in real-time robotic tasks. To address these limitations, we propose YOEO, a single-stage method that simultaneously outputs instance segmentation and NPCS representations in an end-to-end manner. We use a unified network to generate point-wise semantic labels and centroid offsets, allowing points from the same part instance to vote for the same centroid. We further utilize a clustering algorithm to distinguish points based on their estimated centroid distances. Finally, we first separate the NPCS region of each instance. Then, we align the separated regions with the real point cloud to recover the final pose and size. Experimental results on the GAPart dataset demonstrate the pose estimation capabilities of our proposed single-shot method. We also deploy our synthetically-trained model in a real-world setting, providing real-time visual feedback at 200Hz, enabling a physical Kinova robot to interact with unseen articulated objects. This showcases the utility and effectiveness of our proposed method.

Figures

Figures reproduced from arXiv: 2506.05719 by the authors.

Figure 1
Figure 1. Overview. We propose a unified, single-stage method for articulated object 6D pose estimation named YOEO, which enables real-time robotic manipulation. However, there are still significant challenges in perceiving articulated parts. These challenges include: 1) Intra-category part variations. Novel articulated objects often lack exact 3D CAD models, necessitating intra-category generalization. For instance, estimati… view at source ↗
Figure 2
Figure 2. (a) Illustration of challenge posed by intra-category part variations. Estimating the handles of different bucket types [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Architecture overview. The Feature Extraction module extracts the per-point feature from an partial point cloud. They are fed into three parallel modules to predict the NPCS maps, semantic labels and the offsets to centroids of each point. A clustering algorithm is then applied to distinguish different instances with the same semantic label and points on the same instance. Finally, an aligning algorithm is applied t… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The detailed architecture of our YOEO. FC: Fully Connected layer, LFA: Local Feature Aggregation, RS: Random Sampling, MLP: shared Multi-Layer Perceptron, US: Up-sampling. points belonging to the same instance. Part labels are then assigned to each instance to locate t…
Figure 5
Figure 5. Figure 5: Qualitative results on the GAPartNet dataset. The left [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Qualitative results of the real-world perception by [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

47 extracted references · 36 canonical work pages

  1. [1]

    Densefusion: 6d object pose estimation by iterative dense fusion,

    C. Wang, D. Xu, Y . Zhu, R. Mart ´ın-Mart´ın, C. Lu, L. Fei-Fei, and S. Savarese, “Densefusion: 6d object pose estimation by iterative dense fusion,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2019, pp. 3343–3352

    2019

  2. [2]

    I know what you draw: Learning grasp detection conditioned on a few freehand sketches,

    H. Lin, C. Cheang, Y . Fu, and X. Xue, “I know what you draw: Learning grasp detection conditioned on a few freehand sketches,” in 2022 International Conference on Robotics and Automation (ICRA) . IEEE, 2022, pp. 8417–8423

    work page 2022

  3. [3]

    Sar-net: Shape alignment and recovery network for category-level 6d object pose and size estimation,

    H. Lin, Z. Liu, C. Cheang, Y . Fu, G. Guo, and X. Xue, “Sar-net: Shape alignment and recovery network for category-level 6d object pose and size estimation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 6707–6717

    work page 2022

  4. [4]

    Foundationpose: Unified 6d pose estimation and tracking of novel objects,

    B. Wen, W. Yang, J. Kautz, and S. Birchfield, “Foundationpose: Unified 6d pose estimation and tracking of novel objects,” arXiv preprint arXiv:2312.08344, 2023

    arXiv 2023

  5. [5]

    Shapenet: An information-rich 3d model repository,

    A. X. Chang, T. Funkhouser, L. Guibas, P. Hanrahan, Q. Huang, Z. Li, S. Savarese, M. Savva, S. Song, H. Su et al. , “Shapenet: An information-rich 3d model repository,” arXiv preprint arXiv:1512.03012, 2015

    arXiv 2015

  6. [6]

    Garmentnets: Category-level pose estimation for garments via canonical space shape completion,

    C. Chi and S. Song, “Garmentnets: Category-level pose estimation for garments via canonical space shape completion,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2021, pp. 3324–3333

    work page 2021

  7. [7]

    Garmenttracking: Category-level garment pose tracking,

    H. Xue, W. Xu, J. Zhang, T. Tang, Y . Li, W. Du, R. Ye, and C. Lu, “Garmenttracking: Category-level garment pose tracking,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 21 233–21 242

    work page 2023

  8. [8]

    Pourit!: Weakly-supervised liquid percep- tion from a single image for visual closed-loop robotic pouring,

    H. Lin, Y . Fu, and X. Xue, “Pourit!: Weakly-supervised liquid percep- tion from a single image for visual closed-loop robotic pouring,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 241–251

    work page 2023

Show all 47 references
  1. [9]

    Self- supervised transparent liquid segmentation for robotic pouring,

    G. Narasimhan, K. Zhang, B. Eisner, X. Lin, and D. Held, “Self- supervised transparent liquid segmentation for robotic pouring,” in 2022 International Conference on Robotics and Automation (ICRA) . IEEE, 2022, pp. 4555–4561

    2022

  2. [10]

    Toward real-world category-level articulation pose estimation,

    L. Liu, H. Xue, W. Xu, H. Fu, and C. Lu, “Toward real-world category-level articulation pose estimation,” IEEE Transactions on Image Processing, vol. 31, pp. 1072–1083, 2022

    2022

  3. [11]

    Category-level articulated object pose estimation,

    X. Li, H. Wang, L. Yi, L. J. Guibas, A. L. Abbott, and S. Song, “Category-level articulated object pose estimation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 3706–3715

    2020

  4. [12]

    Generative category-level object pose estimation via diffusion models,

    J. Zhang, M. Wu, and H. Dong, “Generative category-level object pose estimation via diffusion models,”Advances in Neural Information Processing Systems, vol. 36, 2024

    2024

  5. [13]

    Fs-net: Fast shape-based network for category-level 6d object pose estimation with decoupled rotation mechanism,

    W. Chen, X. Jia, H. J. Chang, J. Duan, L. Shen, and A. Leonardis, “Fs-net: Fast shape-based network for category-level 6d object pose estimation with decoupled rotation mechanism,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2021, pp....

    2021

  6. [14]

    Gapartnet: Cross-category domain-generalizable object perception and manipulation via generalizable and actionable parts,

    H. Geng, H. Xu, C. Zhao, C. Xu, L. Yi, S. Huang, and H. Wang, “Gapartnet: Cross-category domain-generalizable object perception and manipulation via generalizable and actionable parts,” in Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 202...

    2023

  7. [15]

    Randla-net: Efficient semantic segmentation of large- scale point clouds,

    Q. Hu, B. Yang, L. Xie, S. Rosa, Y . Guo, Z. Wang, N. Trigoni, and A. Markham, “Randla-net: Efficient semantic segmentation of large- scale point clouds,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2020, pp. 11 108–11 117

    2020

  8. [16]

    Least-squares estimation of transformation parameters between two point patterns,

    S. Umeyama, “Least-squares estimation of transformation parameters between two point patterns,” IEEE Transactions on Pattern Analysis & Machine Intelligence , vol. 13, no. 04, pp. 376–380, 1991

    1991

  9. [17]

    Objaverse: A universe of annotated 3d objects,

    M. Deitke, D. Schwenk, J. Salvador, L. Weihs, O. Michel, E. Vander- Bilt, L. Schmidt, K. Ehsani, A. Kembhavi, and A. Farhadi, “Objaverse: A universe of annotated 3d objects,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 13 142–13 153

    2023

  10. [18]

    Objaverse-xl: A universe of 10m+ 3d objects,

    M. Deitke, R. Liu, M. Wallingford, H. Ngo, O. Michel, A. Kusupati, A. Fan, C. Laforte, V . V oleti, S. Y . Gadre et al. , “Objaverse-xl: A universe of 10m+ 3d objects,” Advances in Neural Information Processing Systems, vol. 36, 2024

    2024

  11. [19]

    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

  12. [20]

    Partnet: A large-scale benchmark for fine-grained and hierarchical part-level 3d object understanding,

    K. Mo, S. Zhu, A. X. Chang, L. Yi, S. Tripathi, L. J. Guibas, and H. Su, “Partnet: A large-scale benchmark for fine-grained and hierarchical part-level 3d object understanding,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 909– 918

    2019

  13. [21]

    Deep learning for 3d point clouds: A survey,

    Y . Guo, H. Wang, Q. Hu, H. Liu, L. Liu, and M. Bennamoun, “Deep learning for 3d point clouds: A survey,” IEEE transactions on pattern analysis and machine intelligence , vol. 43, no. 12, pp. 4338–4364, 2020

    2020

  14. [22]

    Google scanned objects: A high-quality dataset of 3d scanned household items,

    L. Downs, A. Francis, N. Koenig, B. Kinman, R. Hickman, K. Rey- mann, T. B. McHugh, and V . Vanhoucke, “Google scanned objects: A high-quality dataset of 3d scanned household items,” in 2022 International Conference on Robotics and Automation (ICRA). IEEE, 2022, pp. 2553–2560

    2022

  15. [23]

    3d compat: Composition of materials on parts of 3d things,

    Y . Li, U. Upadhyay, H. Slim, A. Abdelreheem, A. Prajapati, S. Poth- igara, P. Wonka, and M. Elhoseiny, “3d compat: Composition of materials on parts of 3d things,” in European Conference on Computer Vision. Springer, 2022, pp. 110–127

    2022

  16. [24]

    Neural parts: Learning expressive 3d shape abstractions with invertible neural networks,

    D. Paschalidou, A. Katharopoulos, A. Geiger, and S. Fidler, “Neural parts: Learning expressive 3d shape abstractions with invertible neural networks,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2021, pp. 3204–3215

    2021

  17. [25]

    Partafford: Part-level affordance discovery from 3d objects,

    C. Xu, Y . Chen, H. Wang, S.-C. Zhu, Y . Zhu, and S. Huang, “Partafford: Part-level affordance discovery from 3d objects,” arXiv preprint arXiv:2202.13519, 2022

    2022 arXiv

  18. [26]

    Unsupervised learning for cuboid shape ab- straction via joint segmentation from point clouds,

    K. Yang and X. Chen, “Unsupervised learning for cuboid shape ab- straction via joint segmentation from point clouds,” ACM Transactions on Graphics (TOG) , vol. 40, no. 4, pp. 1–11, 2021

    2021

  19. [27]

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

    C. R. Qi, L. Yi, H. Su, and L. J. Guibas, “Pointnet++: Deep hierarchical feature learning on point sets in a metric space,” Advances in neural information processing systems , vol. 30, 2017

    2017

  20. [28]

    Sgpn: Similarity group proposal network for 3d point cloud instance segmentation,

    W. Wang, R. Yu, Q. Huang, and U. Neumann, “Sgpn: Similarity group proposal network for 3d point cloud instance segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 2569–2578

    2018

  21. [29]

    V otenet: A deep learning label fusion method for multi-atlas segmentation,

    Z. Ding, X. Han, and M. Niethammer, “V otenet: A deep learning label fusion method for multi-atlas segmentation,” in Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I...

    2019

  22. [30]

    3d-sis: 3d semantic instance segmen- tation of rgb-d scans,

    J. Hou, A. Dai, and M. Nießner, “3d-sis: 3d semantic instance segmen- tation of rgb-d scans,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2019, pp. 4421–4430

    2019

  23. [31]

    Pvn3d: A deep point-wise 3d keypoints voting network for 6dof pose estimation,

    Y . He, W. Sun, H. Huang, J. Liu, H. Fan, and J. Sun, “Pvn3d: A deep point-wise 3d keypoints voting network for 6dof pose estimation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 11 632–11 641

    2020

  24. [32]

    Ffb6d: A full flow bidirectional fusion network for 6d pose estimation,

    Y . He, H. Huang, H. Fan, Q. Chen, and J. Sun, “Ffb6d: A full flow bidirectional fusion network for 6d pose estimation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recog- nition, 2021, pp. 3003–3013

    2021

  25. [33]

    Learning 6-dof object poses to grasp category-level objects by language instructions,

    C. Cheang, H. Lin, Y . Fu, and X. Xue, “Learning 6-dof object poses to grasp category-level objects by language instructions,” in 2022 International Conference on Robotics and Automation (ICRA). IEEE, 2022, pp. 8476–8482

    2022

  26. [34]

    Language guided robotic grasping with fine-grained instructions,

    Q. Sun, H. Lin, Y . Fu, Y . Fu, and X. Xue, “Language guided robotic grasping with fine-grained instructions,” in 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2023, pp. 1319–1326

    2023

  27. [35]

    Polaris: Open-ended interactive robotic manipulation via syn2real visual grounding and large language models,

    T. Wang, H. Lin, J. Yu, and Y . Fu, “Polaris: Open-ended interactive robotic manipulation via syn2real visual grounding and large language models,” in 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2024, pp. 9676–9683

    2024

  28. [36]

    Wall-e: Embodied robotic waiter load lifting with large language model,

    T. Wang, Y . Li, H. Lin, X. Xue, and Y . Fu, “Wall-e: Embodied robotic waiter load lifting with large language model,” arXiv preprint arXiv:2308.15962, 2023

    2023 arXiv

  29. [37]

    Normalized object coordinate space for category-level 6d object pose and size estimation,

    H. Wang, S. Sridhar, J. Huang, J. Valentin, S. Song, and L. J. Guibas, “Normalized object coordinate space for category-level 6d object pose and size estimation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2019, pp. 2642–2651

    2019

  30. [38]

    Act the part: Learning interaction strategies for articulated object part discovery,

    S. Y . Gadre, K. Ehsani, and S. Song, “Act the part: Learning interaction strategies for articulated object part discovery,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2021, pp. 15 752–15 761

    2021

  31. [39]

    Akb-48: A real-world articulated object knowledge base,

    L. Liu, W. Xu, H. Fu, S. Qian, Q. Yu, Y . Han, and C. Lu, “Akb-48: A real-world articulated object knowledge base,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 14 809–14 818

    2022

  32. [40]

    Grounding dino: Marrying dino with grounded pre-training for open-set object detection,

    S. Liu, Z. Zeng, T. Ren, F. Li, H. Zhang, J. Yang, C. Li, J. Yang, H. Su, and J. Zhu, “Grounding dino: Marrying dino with grounded pre-training for open-set object detection,” arXiv preprint arXiv:2303.05499, 2023. [Online]. Available: https://arxiv.org/abs/ 2303.05499

    2023 arXiv

  33. [41]

    Shape Prior Deformation for Categorical 6D Object Pose and Size Estimation,

    M. Tian, M. H. Ang, and G. H. Lee, “Shape Prior Deformation for Categorical 6D Object Pose and Size Estimation,” in European Conference on Computer Vision . Springer, 2020, pp. 530–546

    2020

  34. [42]

    Focal loss for dense object detection,

    T.-Y . Lin, P. Goyal, R. Girshick, K. He, and P. Doll ´ar, “Focal loss for dense object detection,” in Proceedings of the IEEE international conference on computer vision , 2017, pp. 2980–2988

    2017

  35. [43]

    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,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2017, pp. 652–660

    2017

  36. [44]

    Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography,

    M. A. Fischler and R. C. Bolles, “Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography,” Communications of the ACM , vol. 24, no. 6, pp. 381–395, 1981

    1981

  37. [45]

    Pointgroup: Dual-set point grouping for 3d instance segmentation,

    L. Jiang, H. Zhao, S. Shi, S. Liu, C.-W. Fu, and J. Jia, “Pointgroup: Dual-set point grouping for 3d instance segmentation,” in Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 4867–4876

    2020

  38. [46]

    Autogpart: Intermediate supervision search for generalizable 3d part segmentation,

    X. Liu, X. Xu, A. Rao, C. Gan, and L. Yi, “Autogpart: Intermediate supervision search for generalizable 3d part segmentation,” inProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 11 624–11 634

    2022

  39. [47]

    Mantisvision camera,

    “Mantisvision camera,” https://www.mantis-vision.com.cn/

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.