Pith. sign in

REVIEW 3 major objections 4 minor 38 references

Exploiting Radiance Fields for Grasp Generation on Novel Synthetic Views

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Novel synthetic views rendered from a Gaussian-splat radiance field add force-closure grasps and improve object coverage beyond what sparse real views provide.

desk verdict A clean, honest initial result that is confounded: adding 16 rendered views alongside 3 real views improves grasps on GraspNet, but the experiment never separates more viewpoints from novel-view synthesis, so the central claim needs a control before it lands. read the letter →

arxiv 2505.11467 v1 pith:S3KZZLDZ submitted 2025-05-16 cs.RO cs.CV

Abhishek Kashyap , Henrik Andreasson , Todor Stoyanov This is my paper · ORCID
classification cs.ROcs.CV
keywords Gaussiansplattingnovelviewsynthesisradiancefieldsgraspgenerationforce-closurecoverageNet-1billionrobotmanipulation
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 tests whether synthetic views rendered from a radiance field can substitute for physically moving a camera to new viewpoints during grasp planning. Using three real RGB-D views of a tabletop scene to build a Gaussian-splatting reconstruction, the authors render sixteen novel viewpoints and run a pretrained grasp detector on all three sets. Across 190 GraspNet-1billion scenes, they find that the novel renders contribute additional force-closure grasps beyond those from real views and raise grasp coverage, meaning the share of objects with at least one grasp. The paper presents these as initial results and stresses that force-closure is an upper-bound metric, not a guarantee that a grasp can be executed.

What carries the argument

The load-bearing object is the radiance-field render: a Gaussian-splatting reconstruction built from sparse RGB-D views that can be projected onto user-specified virtual camera poses to produce color and depth images. Point clouds generated from those renders are fed to a pretrained grasp detector, and each proposed grasp is labelled force-closure using a friction-dependent test. These renders supply viewpoints without moving the robot, which is the mechanism by which additional grasp context is obtained.

What would settle it

Run the novel-view grasps on a real robot, or compare rendered depth at each novel viewpoint against ground-truth depth for the same viewpoint from the GraspNet-1billion captures. If most grasps that appear only in synthetic views fail, or if the rendered point clouds deviate from measured depth beyond grasp tolerance, the extra grasps are artifacts of the reconstruction rather than useful scene context.

Watch

Extended reading notes

Core claim

The paper's central claim is that novel view synthesis adds useful context for grasp generation: a Gaussian-splatting radiance field built from only three real RGB-D views can be rendered from sixteen nearby virtual cameras, and force-closure grasps found on those renders supplement the grasps found on the real views. In the experiments, the union of real and novel views produced more force-closure grasps than real views alone and gave at least one grasp for objects that previously had none, raising per-scene grasp coverage by up to four objects in the GraspNet-1billion benchmark. The authors frame this as initial evidence, noting that additional force-closure grasps need not be executable on a real robot due to reachability, collision, or scene-change issues.

Load-bearing premise

The assumption is that point clouds rendered from the Gaussian-splat reconstruction are faithful enough that force-closure grasps found on them correspond to graspable configurations in the real scene, and the paper's only support for this is aggregate image-quality metrics rather than geometric accuracy or physical trials.

Editorial extensions

If this is right

  • A robot can effectively query many viewpoints without moving its camera, so grasp inference no longer has to wait for a multi-position motion sequence.
  • Objects that are occluded or poorly observed from the physical viewpoints can still produce grasp candidates if a virtual viewpoint reveals them.
  • Grasp coverage, the fraction of objects with at least one force-closure grasp, can increase when real and rendered views are combined.
  • The number of force-closure grasps available after aggregation grows with rendered views, though post-processing such as pose-NMS or clustering changes how many survive.
  • The approach points toward decoupling perception cost, the number of real images, from inference richness, the number of views sampled.

Reading between the lines

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

  • A natural next test is viewpoint selection: choosing novel viewpoints that maximize unseen object surfaces or expected new grasps, rather than fixed nearby offsets, would likely amplify the reported effect.
  • Because the evaluation relies on rendered point clouds, one could separate genuine gains from reconstruction artifacts by comparing rendered depth at a novel viewpoint to ground-truth depth at the same viewpoint before trusting the new grasps.
  • The same pipeline might work with a single real image if combined with a single-image radiance-field or diffusion-based view generator, extending the idea to settings where only one camera snapshot is available.
  • If the effect transfers to real hardware, grasp planners could treat radiance fields as a virtual camera-motion budget: spend physical motion only where virtual views fail.
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

3 major / 4 minor

Summary. The paper investigates whether viewpoint synthesis from a Gaussian-splatting radiance field can improve grasp generation. The authors select three real RGB-D views from each GraspNet-1billion scene, reconstruct the scene with SplaTAM Gaussian Splatting, and render sixteen novel views near the real ones. They then run the pretrained GraspNet-1billion detector on the real views and on the real-plus-novel set, aggregate the grasps, and evaluate them with a Dex-Net 2.0 force-closure check. Their central claim is that novel views add force-closure grasps and increase object-level grasp coverage relative to using only the three real views. The experiments report histogram results across 190 scenes, with coverage gains of up to four objects per scene and a large cluster of scenes gaining roughly 700 additional force-closure grasps.

Significance. If the finding is robust, it is practically useful, because it suggests that a robot can reduce physical camera motions by reconstructing a radiance field from a few viewpoints and then sampling extra synthetic views for grasp inference. The paper's methodology has strengths: it uses established external components (GraspNet-1billion for detection, Dex-Net 2.0 for force-closure, SplaTAM for reconstruction), so the central result is not built on a fitted parameter or a circular definition. The authors also explicitly acknowledge that force-closure is not the same as executable grasping, and they state the need for real-robot verification. However, the empirical support is currently underdetermined by the experimental design because the comparison changes both the number of viewpoints and their source, and because the reconstruction-quality metrics are not shown to be predictive of contact-level geometric fidelity.

major comments (3)
  1. [Sections IV-B and IV-C] The comparison between G_real+nvs and G_real conflates two variables: the number of viewpoints (3 versus 19) and the source of the additional viewpoints (real versus synthesized). The observed increase in force-closure grasps and coverage could result simply from having more viewpoints that provide additional partial observations, independent of whether novel-view synthesis is accurate. To support the claim that NVS specifically provides additional context, the paper should include a control condition that adds 16 real views from the GraspNet-1billion quarter-sphere to the 3 base views, matching the total viewpoint count across conditions.
  2. [Sections III-D and IV-A] The reconstruction-quality metrics in Section IV-A (average PSNR 30.608, MS-SSIM 0.984, LPIPS 0.053, depth L1 0.105) are aggregate image-space scores. They do not establish that the point clouds rendered from the Gaussian-splatting model are geometrically faithful at the contact scale used by the force-closure check in Section III-E. Because grasps are evaluated on rendered point clouds, hallucinated or smoothed geometry could produce force-closure grasps that are not realizable in the original scene. The paper should validate the synthetic-view point clouds against the available dataset geometry, for example by measuring contact-region depth errors or by checking whether synthetic-view grasps are also force-closure when evaluated on real depth from nearby views.
  3. [Section IV-C] The histograms in Figs. 4 and 5 are reported without error bars, confidence intervals, or significance tests over the 190 scenes. The paper states that 'about 17 out of 190 scenes benefit from approximately 700 force-closure grasps,' which suggests the effect is concentrated in a subset of scenes, but no summary statistics or paired statistical test are reported. Since the central claim is an empirical improvement, the paper should report the mean and standard deviation of the per-scene gains and a paired test (e.g., Wilcoxon signed-rank) for both force-closure counts and coverage.
minor comments (4)
  1. [Section III-C] The choice to place novel viewpoints 'close to and with similar orientation to the real views' is described qualitatively; the paper should quantify the distribution of novel-view camera poses relative to the real views, since the spatial layout of the added views is a key factor in the grasp-detection outcome.
  2. [Section III-E] The force-closure criterion 'achieved with any mu' over the set {0.2, 0.4, 0.6, 0.8, 1.0} is very lenient and could include grasps that are force-closure only under high friction. Reporting the distribution of mu values needed for force-closure would strengthen the interpretation of the grasp counts.
  3. [Table II and Figs. 4-5] The results are presented as histograms and a single example scene; providing a numeric table with per-scene force-closure counts and coverage values would make the aggregate results more reproducible.
  4. [Throughout] There are minor formatting issues, such as the missing space in 'SectionIV-A' and inconsistent capitalization of 'GraspNet-1billion' (sometimes 'Graspnet-1billion'), which should be corrected in revision.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the study measures external pretrained detectors and analytic force-closure on a public benchmark, with no fitted parameter renamed as a prediction.

full rationale

The paper's derivation chain is a measurement study rather than a fitted or self-referential derivation. SplaTAM (external implementation) reconstructs a Gaussian-splat scene from M=3 GraspNet-1billion RGB-D frames; the pretrained GraspNet-1billion detector proposes grasps on rendered real and novel views; and the Dex-net 2.0 force-closure check (external analytic metric) labels them. None of these components is fitted to the reported outcome, and no parameter is renamed as a prediction. There is no load-bearing self-citation: the only same-author reference (Stoyanov et al. 2016, ref [28]) is background on SDF representations and is not used to justify the method. The main potential concern is structural: G_real+nvs is constructed as the union of G_real and G_nvs (Section IV-B), so for the unfiltered branch the direction of the coverage comparison is monotone by set inclusion. However, the paper's reported evidence (Table II and the histograms in Figs. 4-5) is based primarily on the pose-NMS and clustering branches, whose nonlinear post-processing makes the gains empirical rather than definitional, and the observed number of additionally contributed grasps and newly covered objects is an empirical measurement of the pretrained detector's output, not a logical consequence of the pipeline definitions. The paper also explicitly flags the real validity limit: force-closure is computed on the reconstructed model's point clouds and 'These results require verification on a real robot.' It further notes in Section IV-C that a high number of force-closure grasps does not guarantee executability on a real robot. These are correctness and validation caveats, not circular reasoning. No circular step can be exhibited in the text, so the circularity score is 0.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

No fitted model parameters or invented physical entities appear. The listed numbers are experimental design choices that affect the magnitude of the reported gains: view counts, the permissive force-closure rule, and post-processing thresholds. The axioms are domain assumptions about reconstruction fidelity and transfer of pretrained detectors; the paper reports aggregate quality metrics but no contact-level validation.

free parameters (4)
  • Number of real views M = 3
    Selected to simulate a minimal eye-in-hand motion scenario. The comparison does not normalize gains per view, so the effect of M is not isolated.
  • Number of novel views N = 16
    Chosen by the authors before evaluation. The magnitude of the reported gain scales with how many extra views are added, and no ablation over N is reported.
  • Friction coefficient acceptance rule = any mu in {0.2, 0.4, 0.6, 0.8, 1.0}
    Section III-E reports force-closure if any friction coefficient gives closure, making the metric permissive and inflating the number of accepted grasps.
  • Post-processing thresholds = NMS: 0.03 m and 15 deg; clustering: 0.05 m and 10 deg
    NMS thresholds are defaults from GraspNet, while clustering thresholds appear hand-chosen in Section IV-B2. Different thresholds would change the counts of merged grasps.
assumptions (3)
  • domain assumption SplaTAM reconstructions from 3 RGB-D views are geometrically accurate enough for grasp inference on rendered views.
    Invoked in Sections III-A and III-D and supported by aggregate color and depth metrics in Section IV-A, but not by contact-level or robot-level validation.
  • domain assumption The GraspNet pretrained grasp detector transfers from real RGB-D images to Gaussian-splatting-rendered images and point clouds.
    The detector is run unchanged on rendered views in Section III-D. Novel views are deliberately placed close to real views in Section III-C to stay near the training distribution, but no calibration or domain adaptation is performed.
  • domain assumption Force-closure is a valid proxy for grasp usefulness in this comparison.
    Force-closure is a geometric condition that ignores reachability, collisions, and execution risk. The authors explicitly list these unmodeled factors in Section IV-C.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Exploiting Radiance Fields for Grasp Generation on Novel Synthetic Views." pith.science (2026). https://pith.science/paper/S3KZZLDZ

@misc{pith2026250511467,
  author       = {Pith},
  title        = {Pith review of: Exploiting Radiance Fields for Grasp Generation on Novel Synthetic Views},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/S3KZZLDZ}},
  note         = {Machine review of arXiv:2505.11467}
}
read the original abstract

Vision based robot manipulation uses cameras to capture one or more images of a scene containing the objects to be manipulated. Taking multiple images can help if any object is occluded from one viewpoint but more visible from another viewpoint. However, the camera has to be moved to a sequence of suitable positions for capturing multiple images, which requires time and may not always be possible, due to reachability constraints. So while additional images can produce more accurate grasp poses due to the extra information available, the time-cost goes up with the number of additional views sampled. Scene representations like Gaussian Splatting are capable of rendering accurate photorealistic virtual images from user-specified novel viewpoints. In this work, we show initial results which indicate that novel view synthesis can provide additional context in generating grasp poses. Our experiments on the Graspnet-1billion dataset show that novel views contributed force-closure grasps in addition to the force-closure grasps obtained from sparsely sampled real views while also improving grasp coverage. In the future we hope this work can be extended to improve grasp extraction from radiance fields constructed with a single input image, using for example diffusion models or generalizable radiance fields.

Figures

Figures reproduced from arXiv: 2505.11467 by the authors.

Figure 1
Figure 1. Example scene reconstruction showing camera poses as frustums, all [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Example real views of a scene for constructing a radiance field [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Clustering and top-grasp filtering C. Evaluation The final resulting grasps are checked for force-closure as described in section III-E. Table II shows grasps for an example scene for post-processing branches pose-NMS and clustering and top-grasp filtering. The original grasps Greal, Gnvs, and Greal+nvs are too many in number resulting in poor visual clarity, and therefore have not been shown [PITH_FULL_IMAGE:figur… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Histogram of force-closure grasps contributed additionally by [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Histogram of number of objects contributed additionally by [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 28 canonical work pages

  1. [1]

    Amodal 3D Reconstruction for Robotic Manipulation via Stability and Connectivity

    William Agnew, Christopher Xie, Aaron Walsman, Oc- tavian Murad, Yubo Wang, Pedro Domingos, and Sid- dhartha Srinivasa. Amodal 3d reconstruction for robotic manipulation via stability and connectivity. InConfer- ence on Robot Learning, pages 1498–1508. PMLR, 2021. URL https://arxiv.org/abs/2009.13146

    work page Pith review arXiv 2021

  2. [2]

    Volumetric Grasping Network: Real-time 6 DOF Grasp Detection in Clutter

    Michel Breyer, Jen Jen Chung, Lionel Ott, Roland Sieg- wart, and Juan Nieto. V olumetric grasping network: Real- time 6 dof grasp detection in clutter. InConference on Robot Learning, pages 1602–1611. PMLR, 2021. URL https://arxiv.org/abs/2101.01132

    work page Pith review arXiv 2021

  3. [4]

    GenNBV: Generalizable Next-Best- View Policy for Active 3D Reconstruction.arXiv preprint arXiv:2402.16174, 2024

    Xiao Chen, Quanyi Li, Tai Wang, Tianfan Xue, and Jiangmiao Pang. GenNBV: Generalizable Next-Best- View Policy for Active 3D Reconstruction.arXiv preprint arXiv:2402.16174, 2024. URL https://arxiv.org/ abs/2402.16174

    arXiv 2024

  4. [5]

    Graspnerf: Multiview-based 6- dof grasp detection for transparent and specular objects using generalizable nerf

    Qiyu Dai, Yan Zhu, Yiran Geng, Ciyu Ruan, Jiazhao Zhang, and He Wang. Graspnerf: Multiview-based 6- dof grasp detection for transparent and specular objects using generalizable nerf. In2023 IEEE International Conference on Robotics and Automation (ICRA), pages 1757–1763. IEEE, 2023. URL https://arxiv.org/abs/2210. 06575

    work page 2023

  5. [6]

    Graspnet-1billion: A large-scale benchmark for general object grasping

    Hao-Shu Fang, Chenxi Wang, Minghao Gou, and Cewu Lu. Graspnet-1billion: A large-scale benchmark for general object grasping. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11444–11453, 2020. URL https://ieeexplore.ieee. org/document/9156992

    work page arXiv 2020

  6. [7]

    Dex-nerf: Using a neural radiance field to grasp transparent objects.arXiv preprint arXiv:2110.14217, 2021

    Jeffrey Ichnowski, Yahav Avigal, Justin Kerr, and Ken Goldberg. Dex-nerf: Using a neural radiance field to grasp transparent objects.arXiv preprint arXiv:2110.14217, 2021. URL https://arxiv.org/abs/2110. 14217

    arXiv 2021

  7. [8]

    Coarse-to-Fine Q-attention: Efficient Learning for Visual Robotic Manipulation via Discretisation

    Stephen James, Kentaro Wada, Tristan Laidlow, and Andrew J Davison. Coarse-to-fine q-attention: Efficient learning for visual robotic manipulation via discretisa- tion. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13739– 13748, 2022. URL https://arxiv.org/abs/2106.12534

    work page Pith review arXiv 2022

  8. [9]

    NViST: In the Wild New View Synthesis from a Single Image with Transformers

    Wonbong Jang and Lourdes Agapito. NViST: In the Wild New View Synthesis from a Single Image with Transformers.arXiv preprint arXiv:2312.08568, 2023. URL https://arxiv.org/abs/2312.08568

    work page Pith review arXiv 2023

Show all 38 references
  1. [10]

    Synergies between affordance and ge- ometry: 6-dof grasp detection via implicit representa- tions.arXiv preprint arXiv:2104.01542, 2021

    Zhenyu Jiang, Yifeng Zhu, Maxwell Svetlik, Kuan Fang, and Yuke Zhu. Synergies between affordance and ge- ometry: 6-dof grasp detection via implicit representa- tions.arXiv preprint arXiv:2104.01542, 2021. URL https://arxiv.org/abs/2104.01542

    2021 arXiv

  2. [11]

    Splatam: Splat, track & map 3d gaussians for dense rgb-d slam.arXiv preprint arXiv:2312.02126, 2023

    Nikhil Keetha, Jay Karhade, Krishna Murthy Jatavallab- hula, Gengshan Yang, Sebastian Scherer, Deva Ramanan, and Jonathon Luiten. Splatam: Splat, track & map 3d gaussians for dense rgb-d slam.arXiv preprint arXiv:2312.02126, 2023. URL https://arxiv.org/abs/2312. 02126

    2023 arXiv

  3. [12]

    3d gaussian splatting for real-time radiance field rendering.ACM Transactions on Graphics, 42(4):1–14, 2023

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering.ACM Transactions on Graphics, 42(4):1–14, 2023. URL https://dl.acm.org/doi/10.1145/ 3592433

    2023

  4. [13]

    Evo-nerf: Evolving nerf for se- quential robot grasping of transparent objects

    Justin Kerr, Letian Fu, Huang Huang, Yahav Avigal, Matthew Tancik, Jeffrey Ichnowski, Angjoo Kanazawa, and Ken Goldberg. Evo-nerf: Evolving nerf for se- quential robot grasping of transparent objects. In6th annual conference on robot learning, 2022. URL https: //proceedings.ml...

    2022

  5. [14]

    Uncertainty guided policy for active robotic 3d reconstruction using neural radiance fields.IEEE Robotics and Automation Letters, 7 (4):12070–12077, 2022

    Soomin Lee, Le Chen, Jiahao Wang, Alexander Liniger, Suryansh Kumar, and Fisher Yu. Uncertainty guided policy for active robotic 3d reconstruction using neural radiance fields.IEEE Robotics and Automation Letters, 7 (4):12070–12077, 2022. URL https://ieeexplore.ieee.org/ docum...

    2022

  6. [15]

    Pointnetgpd: Detecting grasp configurations from point sets

    Hongzhuo Liang, Xiaojian Ma, Shuang Li, Michael Görner, Song Tang, Bin Fang, Fuchun Sun, and Jianwei Zhang. Pointnetgpd: Detecting grasp configurations from point sets. In2019 International Conference on Robotics and Automation (ICRA), pages 3629–3635. IEEE, 2019. URL https://...

    2019

  7. [16]

    Vision transformer for nerf-based view synthesis from a sin- gle input image

    Kai-En Lin, Yen-Chen Lin, Wei-Sheng Lai, Tsung-Yi Lin, Yi-Chang Shih, and Ravi Ramamoorthi. Vision transformer for nerf-based view synthesis from a sin- gle input image. InProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 806–815, 2023. UR...

    2023

  8. [17]

    RGBGrasp: Image-based Object Grasping by Capturing Multiple Views during Robot Arm Movement with Neural Radiance Fields

    Chang Liu, Kejian Shi, Kaichen Zhou, Haoxiao Wang, Jiyao Zhang, and Hao Dong. RGBGrasp: Image-based Object Grasping by Capturing Multiple Views during Robot Arm Movement with Neural Radiance Fields. IEEE Robotics and Automation Letters, 2024. URL https://arxiv.org/abs/2311.16592

    2024 arXiv

  9. [18]

    ManiGaussian: Dynamic Gaussian Splatting for Multi-task Robotic Manipulation

    Guanxing Lu, Shiyi Zhang, Ziwei Wang, Changliu Liu, Jiwen Lu, and Yansong Tang. ManiGaussian: Dynamic Gaussian Splatting for Multi-task Robotic Manipulation. arXiv preprint arXiv:2403.08321, 2024. URL https: //arxiv.org/abs/2403.08321

    2024 arXiv

  10. [19]

    Be- yond top-grasps through scene completion

    Jens Lundell, Francesco Verdoja, and Ville Kyrki. Be- yond top-grasps through scene completion. In2020 IEEE International Conference on Robotics and Au- tomation (ICRA), pages 545–551. IEEE, 2020. URL https://ieeexplore.ieee.org/document/9197320

    2020

  11. [20]

    Dex-net 2.0: Deep learning to plan robust grasps with synthetic point clouds and analytic grasp metrics.arXiv preprint arXiv:1703.09312, 2017

    Jeffrey Mahler, Jacky Liang, Sherdil Niyaz, Michael Laskey, Richard Doan, Xinyu Liu, Juan Aparicio Ojea, and Ken Goldberg. Dex-net 2.0: Deep learning to plan robust grasps with synthetic point clouds and analytic grasp metrics.arXiv preprint arXiv:1703.09312, 2017. URL https:/...

    2017 arXiv

  12. [21]

    Nerf: Representing scenes as neural radiance fields for view synthesis.Communications of the ACM, 65(1): 99–106, 2021

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view synthesis.Communications of the ACM, 65(1): 99–106, 2021. URL https://dl.acm.org/doi/abs/10.1145/ 3503250

    2021

  13. [22]

    6-dof graspnet: Variational grasp generation for object manipulation

    Arsalan Mousavian, Clemens Eppner, and Dieter Fox. 6-dof graspnet: Variational grasp generation for object manipulation. InProceedings of the IEEE/CVF inter- national conference on computer vision, pages 2901– 2910, 2019. URL https://ieeexplore.ieee.org/document/ 9010919

    2019

  14. [23]

    Constructing force-closure grasps

    Van-Duc Nguyen. Constructing force-closure grasps. The International Journal of Robotics Research, 7(3): 3–16, 1988. URL https://ieeexplore.ieee.org/document/ 1087483

    1988

  15. [24]

    Perceiver-actor: A multi-task transformer for robotic ma- nipulation

    Mohit Shridhar, Lucas Manuelli, and Dieter Fox. Perceiver-actor: A multi-task transformer for robotic ma- nipulation. InConference on Robot Learning, pages 785–799. PMLR, 2023. URL https://arxiv.org/abs/2209. 05451

    2023

  16. [25]

    Grasping in the wild: Learning 6dof closed- loop grasping from low-cost demonstrations.IEEE Robotics and Automation Letters, 5(3):4978–4985, 2020

    Shuran Song, Andy Zeng, Johnny Lee, and Thomas Funkhouser. Grasping in the wild: Learning 6dof closed- loop grasping from low-cost demonstrations.IEEE Robotics and Automation Letters, 5(3):4978–4985, 2020. URL https://arxiv.org/abs/1912.04344

    2020 arXiv

  17. [26]

    Gradient based grasp pose optimization on a nerf that approxi- mates grasp success

    Gergely Sóti, Björn Hein, and Christian Wurll. Gradient based grasp pose optimization on a nerf that approxi- mates grasp success. InInternational Conference on In- telligent Autonomous Systems, pages 303–318. Springer,

  18. [27]

    6-DoF Grasp Pose Evaluation and Optimization via Transfer Learning from NeRFs.arXiv preprint arXiv:2401.07935, 2024

    Gergely Sóti, Xi Huang, Christian Wurll, and Björn Hein. 6-DoF Grasp Pose Evaluation and Optimization via Transfer Learning from NeRFs.arXiv preprint arXiv:2401.07935, 2024. URL https://arxiv.org/abs/2401. 07935

    2024 arXiv

  19. [28]

    Grasp envelopes: Extracting constraints on gripper postures from online reconstructed 3d mod- els

    Todor Stoyanov, Robert Krug, Rajkumar Muthusamy, and Ville Kyrki. Grasp envelopes: Extracting constraints on gripper postures from online reconstructed 3d mod- els. In2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 885–892. IEEE, 2016. UR...

    2016

  20. [29]

    Contact-graspnet: Efficient 6- dof grasp generation in cluttered scenes

    Martin Sundermeyer, Arsalan Mousavian, Rudolph Triebel, and Dieter Fox. Contact-graspnet: Efficient 6- dof grasp generation in cluttered scenes. In2021 IEEE International Conference on Robotics and Automation (ICRA), pages 13438–13444. IEEE, 2021. URL https: //ieeexplore.ieee....

    2021

  21. [30]

    Grasp pose detection in point clouds.The International Journal of Robotics Research, 36(13-14): 1455–1473, 2017

    Andreas Ten Pas, Marcus Gualtieri, Kate Saenko, and Robert Platt. Grasp pose detection in point clouds.The International Journal of Robotics Research, 36(13-14): 1455–1473, 2017. URL https://journals.sagepub.com/doi/ 10.1177/0278364917735594

    2017 doi

  22. [32]

    Proba- bilistic 3D multilabel real-time mapping for multi-object manipulation

    Kentaro Wada, Kei Okada, and Masayuki Inaba. Proba- bilistic 3D multilabel real-time mapping for multi-object manipulation. In2017 IEEE/RSJ International Confer- ence on Intelligent Robots and Systems (IROS), pages 5092–5099. IEEE, 2017. URL https://arxiv.org/abs/2001. 05752

    2017

  23. [33]

    Morefusion: Multi- object reasoning for 6d pose estimation from volumetric fusion

    Kentaro Wada, Edgar Sucar, Stephen James, Daniel Lenton, and Andrew J Davison. Morefusion: Multi- object reasoning for 6d pose estimation from volumetric fusion. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 14540– 14549, 2020. URL ...

    2020

  24. [34]

    URL https://arxiv.org/abs/2402.13255

    arXiv

  25. [35]

    pixelnerf: Neural radiance fields from one or few images

    Alex Yu, Vickie Ye, Matthew Tancik, and Angjoo Kanazawa. pixelnerf: Neural radiance fields from one or few images. InProceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 4578–4587, 2021. URL https://ieeexplore.ieee.org/ document/9577688

    2021

  26. [36]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. InProceed- ings of the IEEE conference on computer vision and pattern recognition, pages 586–595, 2018. URL https: //ieeexplo...

    2018

  27. [37]

    Multiscale structural similarity for image quality assess- ment

    Zhou Wang, Eero P Simoncelli, and Alan C Bovik. Multiscale structural similarity for image quality assess- ment. InThe Thrity-Seventh Asilomar Conference on Signals, Systems & Computers, 2003, volume 2, pages 1398–1402. Ieee, 2003. URL https://ieeexplore.ieee.org/ document/1292216

    2003

  28. [41]

    URL https://arxiv.org/abs/2403.09637

    arXiv

  29. [2023]

    URL https://link.springer.com/chapter/10.1007/ 978-3-031-44981-9_26

  30. [2024]

    URL https://arxiv.org/abs/2401.03890

    arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.