Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

Particle-based 6D Object Pose Estimation from Point Clouds using Diffusion Models

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

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

Pith's one-line read A diffusion model trained on SE(3) poses, followed by a cheap particle-selection heuristic, estimates 6D object poses from single-view point clouds at 97.4% accuracy on Linemod.

desk verdict Solid depth-only pose estimator with real engineering, but the generative-multimodality claim rests on an incorrect inference prior that the authors never justify. read the letter →

arxiv 2412.00835 v1 pith:WF34IPIU submitted 2024-12-01 cs.CV

Christian Möller , Niklas Funk , Jan Peters This is my paper · ORCID
classification cs.CV
keywords 6DobjectposeestimationdiffusionmodelspointcloudsSE(3)-equivariantlatentparticleselectionscorematchingLinemod
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

Object pose estimation from a single depth view is ambiguous: occlusion, partial visibility, and object symmetry can make several poses fit the same scene. This paper argues that the right response is to generate many pose hypotheses with a diffusion model trained on SE(3) poses, then choose one with a cheap geometric or score-based heuristic. On the Linemod benchmark the resulting point-cloud-only pipeline reaches a mean ACC-0.1 of 97.4%, and the authors show that the multi-hypothesis step is essential: inferring a single pose cuts performance to roughly 55% AUC versus about 82% with particle selection. If the claim holds, it means pose ambiguity does not need a separately trained ranking network; the generative model plus a simple selection rule is enough.

What carries the argument

The central object is a noise-conditioned score model over the Lie group SE(3): a network that, given a scene point cloud, a partially rendered object point cloud under a current pose hypothesis, and a diffusion time step, predicts a score vector used to refine the pose by Langevin dynamics on the group. Around this sits an SE(3)-equivariant vector-neuron point cloud encoder shared by scene and object, a partial rendering module that keeps only front-facing object points, and the two particle-selection heuristics (by score and by latent). The equivariant latent does double duty: it makes latent-distance selection meaningful and lets the method skip object re-encoding for several Langevin steps, trading about 3% accuracy for an 85% runtime reduction.

What would settle it

Take a held-out set of cluttered scenes or objects not seen in training, sample the usual 20 particles, and compute the Spearman correlation between each selection heuristic's ranking and the ground-truth ADD pose error. If the latent-distance ranking is uncorrelated or anti-correlated with ADD error there, the selection-by-latent contribution fails; if it correlates, the proximity assumption is general rather than a Linemod artifact.

Watch

Extended reading notes

Core claim

The paper's central claim is that a noise-conditioned score model trained by denoising score matching directly on poses in SE(3) can serve as a complete 6D pose estimator from point clouds. At inference, annealed Langevin dynamics draws several pose particles from random initial poses, and the final estimate is picked by one of two training-free rules: the norm of the last predicted score (selection by score) or the Euclidean distance between the SE(3)-equivariant object latent and the scene latent (selection by latent). The authors report 97.4% mean ACC-0.1 on Linemod with ground-truth segmentation masks, above the closest point-cloud baseline CloudAAE+ICP (95.5%) and the RGB-D DenseFusion (94.3%), and they show the selection step is load-bearing: a single particle gives about 55% AUC versus about 82% for score-based selection. They further argue that SE(3) equivariance of the latent space is a core design choice, since removing translation equivariance drops selection-by-latent accuracy from 95.6% to 52.0%.

Load-bearing premise

Selection by latent assumes that the Euclidean distance between the object latent and the scene latent is a reliable monotone proxy for pose correctness, a property no training objective enforces; the paper itself shows in Section 5.3 that the heuristic collapses to 52.0% accuracy when the latent is only SO(3)-equivariant, so the assumption is load-bearing and fragile.

Editorial extensions

If this is right

  • Generative sampling plus selection is a workable alternative to single-shot regression: with one particle the method lands at roughly 55% AUC, so the multimodality is being handled by the sampling process itself.
  • No extra ranking network is needed: both proposed selection heuristics are training-free, and selection by score nearly reaches the oracle upper bound that selects particles by ground-truth error.
  • A single model trained across all 13 objects handles symmetric objects almost perfectly (near 100% on eggbox and glue), supporting the multimodality argument.
  • The SE(3)-equivariant latent enables a direct runtime-accuracy dial: re-rendering the object only every 10th iteration cuts inference time by 85% with a mean accuracy drop of about 3 percentage points.
  • The method stays competitive with RGB and RGB-D approaches while operating only on point clouds, without ICP refinement.

Reading between the lines

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

  • If the latent-proximity assumption transfers, the same object-to-scene latent distance could be reused as a per-hypothesis uncertainty or confidence signal for downstream robotics, since it is already computed during selection.
  • The reported collapse of selection-by-latent under SO(3)-only equivariance indicates the heuristic depends on careful latent geometry; a testable extension is to add an explicit metric or contrastive loss that enforces distance-as-pose-error, which could close the remaining gap to oracle selection.
  • The authors' own observation that selection by score benefits from multi-object training suggests the score magnitude carries a global calibration component; testing on unseen object categories would reveal whether that calibration transfers.
  • The reported 97.4% result uses ground-truth masks, and the paper notes the mean drops to 82.9% with Mask-RCNN masks, so a natural end-to-end test is to couple this estimator with a learned segmentation module and measure the combined gap.
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.

Referee Report

4 major / 6 minor

Summary. The paper proposes a point-cloud-only 6D object pose estimation method based on a score-based diffusion model on SE(3). A Vector Neuron encoder produces SE(3)-equivariant latents for scene and partially rendered object point clouds; during inference, Langevin dynamics refines N randomly initialized pose particles, and one of two proposed selection strategies (score-norm or latent proximity) picks the final pose. On Linemod with ground-truth masks and ADD(S) at 10% of object diameter, the method reports 97.4% mean accuracy using selection by score. The paper includes extensive ablations on particle count, iteration count, rendering interval, partial vs. full rendering, and SE(3) vs. SO(3) equivariance, and releases code.

Significance. If the results hold, the paper is a useful empirical contribution: it demonstrates that multi-hypothesis sampling with a cheap selection heuristic can outperform single-shot pose inference in the point-cloud setting, and that an SE(3)-equivariant latent space can accelerate inference by avoiding per-iteration re-encoding. The manuscript is transparent about its protocol (GT masks), provides detailed implementation and pseudocode, and reports thorough ablations. The two selection strategies are novel in this context and are evaluated with rank-correlation statistics. However, the central generative interpretation and the headline comparison are undermined by the initialization mismatch and the mixed evaluation protocol; these issues require substantive revision.

major comments (4)
  1. [Sec. 3.2, Inference (and Figs. 3, 12)] The initialization H0 = Expmap(ε) with ε ∼ N(0, var(σ,1)I6) is not the marginal q1 of the forward process. For the SDE dx=σ_t dw, q1(Ĥ)=∫ N(Ĥ|H, var(σ,1)) pdata(H)dH, so starting from a zero-mean Gaussian around the identity is only valid if pdata is a point mass at the identity. On Linemod, after the 500× scaling, object translations are typically 1–2 units, while var(σ,1)^(1/2) ≈ 0.51, so the initial particles are several standard deviations away from the training noise-perturbed poses. The score model is trained only inside the noise-perturbed neighborhood of true poses, and no argument is given that its gradients are meaningful at these distant initializations. The large gain from 1 to 20 particles (AUC 0.55 vs 0.82 in Fig. 3; ACC ≈50% vs ≈99% on driller in Fig. 12) is therefore also consistent with random-restart search plus local refinement, and the paper's claim that the diffusion model generates samples from a multimodal posterior (Abstract, Sec. 5.1, Conclusion) is not established. Please either sample the initial particles from a correct prior or a proposal distribution, analyze the behavior of the score far from the data, or substantially temper the generative-sampling claim.
  2. [Sec. 5.4, Table 2] The headline comparison is not apples-to-apples. The reported 97.4% is obtained with ground-truth test masks (acknowledged in the text), whereas most listed baselines do not use ground-truth masks; only CloudAAE is stated to share the GT-mask protocol. The sentence "we surpass the performance of DenseFusion by 3.1 percentage points" is therefore not supported by the table as presented. The paper itself reports that with Mask-RCNN masks the mean accuracy drops to 82.9%, below the OVE6D number of 86.1% quoted in the same paragraph. To support the competitive claim, the main table should separate GT-mask from automatic-mask protocols, or the comparison should be re-run under a unified detection/mask setting.
  3. [Appendix Sec. 11.1] Selection-by-score is not robust when training a single-object model: the reported ACC-0.1 drops from 97.4% to 72.7% for the driller, while selection-by-latent remains unaffected. Since selection-by-score is the strategy used for all headline results, this is a load-bearing limitation, not a minor ablation. It suggests that the score norm is not a universal quality measure and that the main-table numbers may depend on side effects of multi-object training. The main text should either analyze this effect or explicitly state that selection-by-score is only reliable in the multi-object training setting, rather than presenting it as generally effective.
  4. [Sec. 3.3 and Sec. 5.3] The selection-by-latent strategy rests on an unenforced assumption: Euclidean distance in the SE(3)-equivariant latent space should be a monotone proxy for pose error. The encoder is trained solely with the score-matching loss (Sec. 3.2, Eq. 1), which imposes no such metric property. The fragility is documented in Sec. 5.3: with an SO(3)-equivariant latent, accuracy drops from 95.6% to 52.0%. This does not invalidate the empirical result on Linemod, but it means that transfer to cluttered scenes, novel objects, or different depth distributions is not supported by the current evidence. At minimum, the paper should state this assumption explicitly and provide a direct correlation analysis between l_prox and ADD, not only rank correlations among sampled particles.
minor comments (6)
  1. [Sec. 3, first paragraph] "We proposes learning" should be "We propose learning".
  2. [Algorithm 1, line 4] The pseudocode condition "if l mod k" should read "if l mod k == 0" (or equivalent); as written, for k=1 the condition is always false, so the object point cloud is never rendered in the pseudocode, which contradicts the described behavior.
  3. [Sec. 5.5 and Table 3] The text reports an accuracy drop of "-3%" for k=10, but Table 3 reports -2.8 percentage points; please harmonize the numbers.
  4. [Fig. 11] The y-axis label "ACC-0.1 [%]" shows values between 0.2 and 1.0; either the label should be a fraction or the values should be multiplied by 100.
  5. [Sec. 3.3] In the derivation for selection-by-score, the sign convention and the time-dependent scaling of the score target should be stated explicitly; the text says the score "should match ∇_Ĥ log q_t(Ĥ|H) ∝ Logmap(H^{-1}Ĥ)" without specifying the sign, and the Euclidean analogy omits the 1/var(σ,t) factor used in Eq. (1).
  6. [Sec. 5.4] OVE6D is discussed as a close baseline but does not appear in Table 2; including it would make the comparison easier to follow.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the particle-selection claims are validated empirically against ground-truth rankings, and prior-work citations are methodological, not load-bearing.

full rationale

The derivation chain is self-contained. The diffusion training objective (Eq. 1) and Langevin inference (Sec. 3.2) are standard score-matching recipes, and the paper's novel contributions, the two particle-selection strategies, are not derived from a fitted target but are proposed as heuristics and tested against ground-truth particle rankings (Fig. 4, Tab. 1). Selection-by-score uses the trained score model's norm; selection-by-latent uses the equivariant encoder's distance. Neither property is baked in by a loss that optimizes the selection metric: the encoder is trained only by denoising score matching, and the paper itself shows the latent-proximity assumption fails when the latent is only SO(3)-equivariant (Sec. 5.3), which confirms the heuristic has empirical content. References to prior SE(3) diffusion work [36] (co-authored by two of the present authors) supply the diffusion framework and time encoding, but the central pose-estimation and particle-selection claims are evaluated against Linemod and compared with external baselines, so the self-citations are not load-bearing. The initialization at N(0,var(σ,1)) rather than the true prior q1 is a potential sampling or correctness concern, not a circularity: it does not make the reported results equal to the inputs by construction.

Assumptions & free parameters 8 free parameters · 6 assumptions · 0 invented entities

The method does not introduce new physical entities. The free parameters are standard deep-learning hyperparameters chosen by hand or tuned on the driller instance. The axioms are a mix of standard diffusion theory, geometric properties of equivariant networks, and two ad hoc heuristics that form the basis of the particle selection strategies. The latter are the most fragile assumptions in the paper.

free parameters (8)
  • Latent dimension D = 94
    Hand-chosen dimension of the SE(3)-equivariant latent space for both encoders.
  • Noise scale sigma = 0.5
    Controls the perturbation schedule of the diffusion SDE (Sec. 3.2).
  • Number of Langevin iterations L = 100
    Set as default after ablations on the driller object (Appendix 11.3); accuracy saturates after 50.
  • Number of particles N = 20
    Selected after ablation on driller; accuracy saturates at 20.
  • Rendering interval k = 1 (default), 10 (fast)
    Controls how often the object point cloud is re-rendered and re-encoded; tuned on driller and evaluated across all objects.
  • Langevin noise coefficient = 0.01
    Scales the exploration noise in the Langevin update (Sec. 3.2).
  • Partial-rendering threshold distribution = N(0, 0.1), clipped to [-0.3, 0.3]
    Random front-facing margin sampled per forward pass to robustify the visibility heuristic (Sec. 3.1).
  • Step size formula alpha_l = 2 / ||s_l||^2
    Hand-specified step-size rule in the Langevin update (Algorithm 1), not derived from the SDE discretization.
assumptions (6)
  • standard math Score matching and annealed Langevin dynamics provide valid generative sampling on SE(3) when poses are parameterized in R6 via the logarithmic map.
    Invoked in Sec. 3.2 and follows Song et al. [30,31] and Urain et al. [36].
  • domain assumption The VN encoder, after centering and mean re-addition, yields a latent space that is SE(3)-equivariant.
    Stated in Sec. 3.1 and Appendix 9, based on Deng et al. [7] and Katzir et al. [19].
  • domain assumption The object 3D model is known and rigid, and its pose is fully described by a single SE(3) transform.
    Assumed throughout; needed for partial rendering and point sampling.
  • domain assumption The front-facing normal heuristic is a sufficient approximation of partial observability from a depth view.
    Introduced in Sec. 3.1 as a heuristic; occlusions by other objects are explicitly not modeled.
  • ad hoc to paper The Euclidean distance between object and scene latents is a meaningful proxy for pose accuracy for the selection-by-latent strategy.
    Modeling assumption central to one of the two selection strategies; not enforced by training and shown to be fragile for SO(3)-only equivariance.
  • ad hoc to paper The norm of the final score is a meaningful quality measure for a pose hypothesis for the selection-by-score strategy.
    Heuristic introduced in Sec. 3.3, motivated by score matching but not rigorously proven; validated empirically.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Particle-based 6D Object Pose Estimation from Point Clouds using Diffusion Models." pith.science (2026). https://pith.science/paper/WF34IPIU

@misc{pith2026241200835,
  author       = {Pith},
  title        = {Pith review of: Particle-based 6D Object Pose Estimation from Point Clouds using Diffusion Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WF34IPIU}},
  note         = {Machine review of arXiv:2412.00835}
}
read the original abstract

Object pose estimation from a single view remains a challenging problem. In particular, partial observability, occlusions, and object symmetries eventually result in pose ambiguity. To account for this multimodality, this work proposes training a diffusion-based generative model for 6D object pose estimation. During inference, the trained generative model allows for sampling multiple particles, i.e., pose hypotheses. To distill this information into a single pose estimate, we propose two novel and effective pose selection strategies that do not require any additional training or computationally intensive operations. Moreover, while many existing methods for pose estimation primarily focus on the image domain and only incorporate depth information for final pose refinement, our model solely operates on point cloud data. The model thereby leverages recent advancements in point cloud processing and operates upon an SE(3)-equivariant latent space that forms the basis for the particle selection strategies and allows for improved inference times. Our thorough experimental results demonstrate the competitive performance of our approach on the Linemod dataset and showcase the effectiveness of our design choices. Code is available at https://github.com/zitronian/6DPoseDiffusion .

Figures

Figures reproduced from arXiv: 2412.00835 by the authors.

Figure 1
Figure 1. Noise Conditioned Score Model (NCSM) architecture [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Objects of the Linemod dataset [13] excluding the two objects with incomplete 3D models. * denotes symmetric objects. ues {s0, s1, · · · , sL−1} ∈ R L×6 . As shown in [30], for score matching in Euclidean Spaces, i.e., with qt(xˆ|x) = N (xˆ|x, σt), the objective for score matching equates to ∇xˆlog(qt(xˆ|x)) = −(xˆ − x) 2/σ2 t . For our case of score matching in SE(3), the Gaussian is defined according to qt(Hˆ |H, … view at source ↗
Figure 3
Figure 3. Accuracy Curve with AUC values for all selection meth [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Heatmap showing the correlation between the true rank [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 6
Figure 6. Figure 6: Comparison of using a SE(3)-equivariant latent with its SO(3)-equivariant alternative. (a) is based on the particle selection by score and (b) based on the selection by latent. Approach PVNet [24] PoseCNN + DeepIm [22, 39] DenseFusion [37] HybridPose [29] CloudAAE + IC…
Figure 7
Figure 7. Figure 7: Relation between the rendering interval k, accuracy, and inference runtime. The interval k controls how often the point cloud is rendered and encoded during inference; in other iterations, only the latent vector zo is transformed. Results are based on 100 iterations, 2…
Figure 8
Figure 8. Figure 8: Geometrical visualization of the process to determines [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: Point Cloud encoder implemented through vector neurons [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Comparison of using the partially observable object [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]
Figure 12
Figure 12. Figure 12: Relation between number of sampled particles and pose [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Diffusion-based Inverse Observation Model for Artificial Skin

    cs.RO 2025-06 conditional novelty 5.0 of 10

    A conditional denoising diffusion model learns the inverse tactile observation model of an artificial skin, and in simulation it generates object pose hypotheses that improve particle filter convergence.

Reference graph

Works this paper leans on

47 extracted references · 42 canonical work pages · cited by 1 Pith paper

  1. [1]

    Uncertainty-driven 6d pose estimation of objects and scenes from a single rgb image

    Eric Brachmann, Frank Michel, Alexander Krull, Michael Ying Yang, Stefan Gumhold, and Carsten Rother. Uncertainty-driven 6d pose estimation of objects and scenes from a single rgb image. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 3364–3372, 2016. 5

    work page 2016

  2. [2]

    Efficientpose: An efficient, accurate and scalable end-to-end 6d multi object pose estimation approach, 2020

    Yannick Bukschat and Marcus Vetter. Efficientpose: An efficient, accurate and scalable end-to-end 6d multi object pose estimation approach, 2020. 5

    work page 2020

  3. [3]

    Ove6d: Object viewpoint encoding for depth-based 6d object pose estimation

    Dingding Cai, Janne Heikkil¨a, and Esa Rahtu. Ove6d: Object viewpoint encoding for depth-based 6d object pose estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6803–6813, 2022. 2, 8

    work page 2022

  4. [4]

    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,

    arXiv

  5. [5]

    Equivariant point network for 3d point cloud analysis

    Haiwei Chen, Shichen Liu, Weikai Chen, Hao Li, and Randall Hill. Equivariant point network for 3d point cloud analysis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14514–14523, 2021. 2

    work page 2021

  6. [6]

    Gaussian approxi- mation of non-linear measurement models on lie groups

    Gregory Chirikjian and Marin Kobilarov. Gaussian approxi- mation of non-linear measurement models on lie groups. In IEEE Conference on Decision and Control, 2014. 5

    work page 2014

  7. [7]

    C. Deng, O. Litany, Y . Duan, A. Poulenard, A. Tagliasacchi, and L. Guibas. Vector neurons: A general framework for so(3)-equivariant networks. In 2021 IEEE/CVF International Conference on Computer Vision (ICCV), pages 12180–12189, Los Alamitos, CA, USA, 2021. IEEE Computer Society. 1, 2, 4, 11

    work page 2021

  8. [8]

    Manif: A micro lie theory library for state estimation in robotics applications

    J´er´emie Deray and Joan Sol `a. Manif: A micro lie theory library for state estimation in robotics applications. Journal of Open Source Software, 5:1371, 2020. 4

    work page 2020

Show all 47 references
  1. [9]

    Fuchs, Daniel E

    Fabian B. Fuchs, Daniel E. Worrall, V olker Fischer, and Max Welling. Se(3)-transformers: 3d roto-translation equivariant attention networks. In Proceedings of the 34th International Conference on Neural Information Processing Systems, Red Hook, NY , USA, 2020. Curran Associat...

    2020

  2. [10]

    Cloudaae: Learning 6d object pose regression with on-line data synthesis on point clouds

    Ge Gao, Mikko Lauri, Xiaolin Hu, Jianwei Zhang, and Si- mone Frintrop. Cloudaae: Learning 6d object pose regression with on-line data synthesis on point clouds. In 2021 IEEE In- ternational Conference on Robotics and Automation (ICRA), pages 11081–11087. IEEE, 2021. 2, 7, 8

    2021

  3. [11]

    Are we ready for autonomous driving? the kitti vision benchmark suite

    Andreas Geiger, Philip Lenz, and Raquel Urtasun. Are we ready for autonomous driving? the kitti vision benchmark suite. In 2012 IEEE Conference on Computer Vision and Pattern Recognition, pages 3354–3361, 2012. 1

    2012

  4. [12]

    Model based training, detection and pose estimation of texture-less 3d objects in heavily cluttered scenes

    Stefan Hinterstoisser, Vincent Lepetit, Slobodan Ilic, Ste- fan Holzer, Gary Bradski, Kurt Konolige, and Nassir Navab. Model based training, detection and pose estimation of texture-less 3d objects in heavily cluttered scenes. In Com- puter Vision – ACCV 2012, pages 548–562, B...

    2012

  5. [13]

    Hinterstoisser, S

    S. Hinterstoisser, S. Holzer, C. Cagniart, S. Ilic, K. Konolige, N. Navab, and V . Lepetit. Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes. In IEEE International Conference on Computer Vision (ICCV),

  6. [14]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. In Advances in Neural Information Processing Systems, pages 6840–6851. Curran Associates, Inc., 2020. 2

    2020

  7. [15]

    A comprehensive review on 3d object detection and 6d pose estimation with deep learning

    Sabera Hoque, Md Yasir Arafat, Shuxiang Xu, Ananda Maiti, and Yuchen Wei. A comprehensive review on 3d object detection and 6d pose estimation with deep learning. IEEE Access, 9:143746–143770, 2021. 1, 2

    2021

  8. [16]

    Confronting ambiguity in 6d object pose esti- mation via score-based diffusion on se(3), 2023

    Tsu-Ching Hsiao, Hao-Wei Chen, Hsuan-Kung Yang, and Chun-Yi Lee. Confronting ambiguity in 6d object pose esti- mation via score-based diffusion on se(3), 2023. 1, 2

    2023

  9. [17]

    Se (3) diffusion model-based point cloud registra- tion for robust 6d object pose estimation

    Haobo Jiang, Mathieu Salzmann, Zheng Dang, Jin Xie, and Jian Yang. Se (3) diffusion model-based point cloud registra- tion for robust 6d object pose estimation. Advances in Neural Information Processing Systems, 36, 2024. 2

    2024

  10. [18]

    Yolo-6d+: Single shot 6d pose estimation using privileged silhouette information

    Jia Kang, Wenjun Liu, Wenzhe Tu, and Lu Yang. Yolo-6d+: Single shot 6d pose estimation using privileged silhouette information. In 2020 International Conference on Image Processing and Robotics (ICIP), pages 1–6, 2020. 2, 5

    2020

  11. [19]

    Shape- pose disentanglement using se(3)-equivariant vector neurons

    Oren Katzir, Dani Lischinski, and Daniel Cohen-Or. Shape- pose disentanglement using se(3)-equivariant vector neurons. In Computer Vision - ECCV 2022 - 17th European Conference, Tel Aviv, Israel, October 23-27, 2022, Proceedings, Part III, pages 468–484. Springer, 2022. 2

    2022

  12. [20]

    A tutorial on energy-based learning

    Yann Lecun, Sumit Chopra, Raia Hadsell, Marc Aurelio Ran- zato, and Fu Jie Huang. A tutorial on energy-based learning. MIT Press, 2006. 1

    2006

  13. [21]

    Generative 9 category-level shape and pose estimation with semantic prim- itives

    Guanglin Li, Yifeng Li, Zhichao Ye, Qihang Zhang, Tao Kong, Zhaopeng Cui, and Guofeng Zhang. Generative 9 category-level shape and pose estimation with semantic prim- itives. In 6th Annual Conference on Robot Learning, 2022. 2, 3

    2022

  14. [22]

    Deepim: Deep iterative matching for 6d pose estimation

    Yi Li, Gu Wang, Xiangyang Ji, Yu Xiang, and Dieter Fox. Deepim: Deep iterative matching for 6d pose estimation. International Journal of Computer Vision, 128(3):657–678,

  15. [23]

    Pose estimation for augmented reality: A hands-on survey

    Eric Marchand, Hideaki Uchiyama, and Fabien Spindler. Pose estimation for augmented reality: A hands-on survey. IEEE Transactions on Visualization and Computer Graphics , 22 (12):2633–2651, 2016. 1

    2016

  16. [24]

    Pvnet: Pixel-wise voting network for 6dof pose estimation

    Sida Peng, Yuan Liu, Qixing Huang, Xiaowei Zhou, and Hujun Bao. Pvnet: Pixel-wise voting network for 6dof pose estimation. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4556–4565,

    2019

  17. [25]

    Charles Ruizhongtai Qi, Hao Su, Kaichun Mo, and Leonidas J. Guibas. Pointnet: Deep learning on point sets for 3d classi- fication and segmentation. CoRR, abs/1612.00593, 2016. 1, 11

    2016 arXiv

  18. [26]

    Qi, Li Yi, Hao Su, and Leonidas J

    Charles R. Qi, Li Yi, Hao Su, and Leonidas J. Guibas. Point- net++: Deep hierarchical feature learning on point sets in a metric space. In Proceedings of the 31st International Conference on Neural Information Processing Systems, page 5105–5114, Red Hook, NY , USA, 2017. Curra...

    2017

  19. [27]

    Shelving, stacking, hanging: Relational pose diffusion for multi-modal rearrangement

    Anthony Simeonov, Ankit Goyal, Lucas Manuelli, Lin Yen- Chen, Alina Sarmiento, Alberto Rodriguez, Pulkit Agrawal, and Dieter Fox. Shelving, stacking, hanging: Relational pose diffusion for multi-modal rearrangement. Conference on Robot Learning, 2023. 1, 2

    2023

  20. [28]

    A micro lie theory for state estimation in robotics

    Joan Sola, Jeremie Deray, and Dinesh Atchuthan. A micro lie theory for state estimation in robotics. arXiv preprint arXiv:1812.01537, 2018. 4

    2018 arXiv

  21. [29]

    C. Song, J. Song, and Q. Huang. Hybridpose: 6d object pose estimation under hybrid representations. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 428–437, Los Alamitos, CA, USA, 2020. IEEE Computer Society. 1, 2, 5, 7

    2020

  22. [30]

    Generative modeling by estimating gradients of the data distribution

    Yang Song and Stefano Ermon. Generative modeling by estimating gradients of the data distribution. In Advances in Neural Information Processing Systems. Curran Associates, Inc., 2019. 1, 2, 5

    2019

  23. [31]

    Improved techniques for train- ing score-based generative models

    Yang Song and Stefano Ermon. Improved techniques for train- ing score-based generative models. InProceedings of the 34th International Conference on Neural Information Processing Systems, Red Hook, NY , USA, 2020. Curran Associates Inc. 2, 4, 12

    2020

  24. [32]

    Yang Song and Diederik P. Kingma. How to train your energy- based models, 2021. 1

    2021

  25. [33]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Ab- hishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. In International Conference on Learning Representations ,

  26. [34]

    Qi, Jean-Emmanuel Deschaud, Beatriz Marcotegui, Fran c ¸ois Goulette, and Leonidas J

    Hugues Thomas, Charles R. Qi, Jean-Emmanuel Deschaud, Beatriz Marcotegui, Fran c ¸ois Goulette, and Leonidas J. Guibas. Kpconv: Flexible and deformable convolution for point clouds. Proceedings of the IEEE International Confer- ence on Computer Vision, 2019. 2

    2019

  27. [35]

    Smidt, Steven Kearnes, Lusann Yang, Li Li, Kai Kohlhoff, and Patrick Riley

    Nathaniel Thomas, Tess E. Smidt, Steven Kearnes, Lusann Yang, Li Li, Kai Kohlhoff, and Patrick Riley. Tensor field net- works: Rotation- and translation-equivariant neural networks for 3d point clouds. CoRR, abs/1802.08219, 2018. 2

    2018 arXiv

  28. [36]

    Se(3)-diffusionfields: Learning smooth cost functions for joint grasp and motion optimization through diffusion

    Julen Urain, Niklas Funk, Jan Peters, and Georgia Chal- vatzaki. Se(3)-diffusionfields: Learning smooth cost functions for joint grasp and motion optimization through diffusion. IEEE International Conference on Robotics and Automation (ICRA), 2023. 1, 2, 3, 4, 11

    2023

  29. [37]

    C. Wang, D. Xu, Y . Zhu, R. Martin-Martin, C. Lu, L. Fei-Fei, and S. Savarese. Densefusion: 6d object pose estimation by iterative dense fusion. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 3338–3347, Los Alamitos, CA, USA, 2019. IEEE Co...

    2019

  30. [38]

    Sarma, Michael M

    Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E. Sarma, Michael M. Bronstein, and Justin M. Solomon. Dynamic graph cnn for learning on point clouds. ACM Trans. Graph., 38(5), 2019. 1, 2, 11

    2019

  31. [39]

    Posecnn: A convolutional neural network for 6d object pose estimation in cluttered scenes

    Yu Xiang, Tanner Schmidt, Venkatraman Narayanan, and Dieter Fox. Posecnn: A convolutional neural network for 6d object pose estimation in cluttered scenes. 2018. 1, 2, 5, 7

    2018

  32. [40]

    6d-diff: A keypoint diffusion framework for 6d object pose estimation

    Li Xu, Haoxuan Qu, Yujun Cai, and Jun Liu. 6d-diff: A keypoint diffusion framework for 6d object pose estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9676–9686, 2024. 2

    2024

  33. [41]

    Iterative Closest Point (ICP), pages 433–

    Zhengyou Zhang. Iterative Closest Point (ICP), pages 433–

  34. [42]

    E2pn: Efficient se (3)-equivariant point network

    Minghan Zhu, Maani Ghaffari, William A Clark, and Huei Peng. E2pn: Efficient se (3)-equivariant point network. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 1223–1232, 2023. 2 10 Particle-based 6D Object Pose Estimation from Poin...

    2023

  35. [44]

    8 visualizing the core idea behind the partial rendering module that has been presented in Sec

    Additional Information on the Partial Ren- dering Module We provide Fig. 8 visualizing the core idea behind the partial rendering module that has been presented in Sec. 3.1

  36. [45]

    Additional Information on the Point Cloud Encoder The point cloud encoder leverages the SO(3)-equivariant vector neurons (VNs) proposed in [ 7] and defines Pθ : RN ×3 → RD×3 with D being the latent dimension (Fig. 9). We use the encoder architecture proposed for point cloud re...

  37. [46]

    Algorithm 1

    Inference Process - Pseudo Code This section provides the pseudo code for our inference procedure, i.e., cf. Algorithm 1

  38. [47]

    Selection Methods when Training a Single Ob- ject Model While Sec

    Additional Experimental Results 11.1. Selection Methods when Training a Single Ob- ject Model While Sec. 5.1 compared the two selection strategies when training a single model for object pose estimation with all of the objects, this section provides additional results for the ...

  39. [434]

    Springer US, Boston, MA, 2014. 2

    2014

Pith tools

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