Pith. sign in

REVIEW 4 major objections 6 minor 79 references

Generalizable Single-view Object Pose Estimation by Two-side Generating and Matching

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 The paper claims that estimating object pose by generating and matching intermediate views from both the reference and query images, rather than generating the query view directly, substantially improves single-image pose estimation…

desk verdict Two-side matching with score distillation is a plausible extension that deserves serious review, but test-set-tuned hyperparameters and an unvalidated surrogate loss mean the numbers should be treated with caution. read the letter →

arxiv 2411.15860 v1 pith:5MCI2RKP submitted 2024-11-24 cs.CV

Yujing Sun , Caiyi Sun , Yuan Liu , Yuexin Ma , Siu Ming Yiu This is my paper · ORCID
classification cs.CV
keywords objectposeestimationsingle-viewdiffusionmodeltwo-sidematchingnovelviewgenerationscoredistillationcategory-agnosticlargeviewpointchange
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 presents a method to estimate an object's pose from a single RGB image, using only one reference image of the same object. The method uses a pre-trained diffusion model to generate novel views, but instead of generating the query view directly from the reference, it generates intermediate views from both images and matches them on a fixed set of intermediate viewpoints. The two-sided matching reduces the generation burden when the viewpoint change is large, and the paper reports that it achieves the best rotation accuracy on both the synthetic GSO and real NAVI benchmarks, outperforming the strongest prior diffusion-based method by about 13 to 19 percentage points at the 15-degree threshold. No training on the object, no CAD model, and no multiple views are required.

What carries the argument

The key machinery is the two-side matching objective and its score-function approximation. The paper replaces the direct generation of the query view from the reference (which degrades at large viewpoint changes) with a set of intermediate viewpoints sampled evenly on the upper hemisphere via the Fibonacci sphere algorithm. For each intermediate viewpoint, it generates an image from the reference and, conceptually, from the query, and matches them. Because generating both sets of images is costly, the matching distance is approximated by the noise-prediction loss of the pre-trained diffusion model, as in DreamFusion: one adds noise to the intermediate image generated from the reference and asks the denoiser, conditioned on the query image and the relative viewpoint change, to predict that noise; the L2 residual is the matching score. A coarse-to-fine search over azimuth and elevation, followed by gradient descent refinement, finds the pose that minimizes the score.

What would settle it

A direct test: take a set of objects with known 3D models, render a reference view and a query view at a known relative pose, run the method, and record the score function values across the coarse search grid. If the minimum of the score function does not coincide with the true pose for a substantial fraction of test cases, or if the score is flat or multi-modal near the true pose, the score proxy is biased and the central claim would fail. This test can be run offline without any retraining.

Watch

Extended reading notes

Core claim

The central discovery is that a two-side generating-and-matching scheme makes diffusion-based, category-agnostic object pose estimation accurate even for large viewpoint changes. Given a reference image with known pose and a query image of the same object, the method first generates images of the object from N intermediate viewpoints using a pre-trained novel-view diffusion model, once conditioned on the reference and once on the query. It then searches over the query pose by minimizing a score function that approximates the matching distance between these two generated image sets; the score function is the DreamFusion-style score distillation loss between the denoiser's noise prediction and the added noise. The paper reports that this approach achieves the best rotation accuracy on both GSO and NAVI, and that under viewpoint changes of 120 degrees or more it roughly doubles the accuracy of the strongest prior method.

Load-bearing premise

The load-bearing premise is that the DreamFusion score distillation loss used in Eq. (4) is a valid proxy for the two-side image matching objective: the paper assumes that minimizing the noise-prediction residual is equivalent to finding the pose that makes the generated image sets match, without proving that the loss is minimized at the correct pose.

Editorial extensions

If this is right

  • A single reference image and no 3D model or multi-view capture suffice, so the method can be applied to arbitrary novel objects at inference time.
  • Large viewpoint changes, the failure case of prior diffusion pose estimators, become tractable because the matching is done on small, well-conditioned steps.
  • The pipeline is modular with respect to the generative prior: improving the novel-view diffusion model should directly raise pose accuracy.
  • The coarse-to-fine search and refinement are parallelizable, and the paper's light variant processes a query in about one second on a single GPU.
  • Accurate single-image pose estimates enable downstream applications such as augmented-reality object insertion, 3D reconstruction, and robot manipulation from a single photo.

Reading between the lines

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

  • The approximation in Eq. (4) is not proven to have its minimum at the true pose; if the noise-prediction residual is systematically biased, the pose search could be wrong even when intermediate views align. This could be tested by evaluating the score function at ground-truth poses against nearby poses on a large set of known objects.
  • The intermediate viewpoints are fixed and object-independent; adapting them to an object's shape or refining them iteratively might improve accuracy or reduce the number of generated views needed.
  • The method assumes a single centered object in both images; extending to scenes or occluded objects would require detection or segmentation, which the paper only sketches.
  • The reported gains are on two datasets with specific object distributions; broader testing across more diverse objects, lighting, and backgrounds would clarify how far the two-side matching advantage extends.
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 / 6 minor

Summary. The paper proposes a generalizable single-view object pose estimation method, Gen2SM, that uses a pre-trained diffusion model (Zero123) to estimate the relative elevation and azimuth between a reference image and a query image. Instead of directly generating the query view from the reference, the method generates images at N intermediate viewpoints from the reference side and matches them against the query image through a DreamFusion-style score distillation loss, followed by coarse-to-fine grid search and gradient refinement. Experiments on the GSO and NAVI datasets report large improvements over prior methods, especially for large viewpoint changes, along with ablations on the number of intermediate views, Monte-Carlo samples, time step, and refinement iterations.

Significance. If the reported gains are sound, the two-side matching idea is a meaningful step for training-free, category-agnostic pose estimation and could benefit downstream applications such as AR and 3D reconstruction. The paper includes extensive ablations, a runtime analysis, visualizations, and a promise of code release. However, the central loss in Eq. (5) is a heuristic surrogate whose equivalence to the stated two-sided matching objective is not established, and the key hyperparameters are selected on the test set, so the significance of the empirical claims is currently conditional.

major comments (4)
  1. [§3.3, Eqs. (4)–(5)] The replacement of the two-sided matching objective in Eq. (3) by the DreamFusion-style score distillation loss in Eq. (4) is load-bearing but not justified. Eq. (5) is asymmetric: images are generated only from the reference side, and the query image is used as a conditioning signal for denoising reference-side generated views. There is no derivation showing that the expected denoising residual is minimized at the true (θ_q, ϕ_q), nor an empirical demonstration that the global minimum on the search grid is separated from wrong-pose minima. Please provide either a proof or a quantitative validation on held-out objects, e.g., a plot of f(θ_q, ϕ_q) against pose error or a comparison with directly optimizing Eq. (3) on a subset of the data.
  2. [§4.1 and Tables 3–5] All core hyperparameters (N=64, M=4, t=0.4, and 3 refinement iterations) are selected on the NAVI test set, and the paper reports no error bars, per-object variance, or significance tests. Since Table 5 shows a drop from 45.58% to 35.33% at 15° when t changes from 0.4 to 0.8, the reported margin over E2VG could partly reflect test-set selection. Please report results on a validation split or with confidence intervals across objects, and state the exact hyperparameter selection procedure.
  3. [§4.1–4.2, Table 1] The evaluation uses the benchmark constructed in the authors' prior work E2VG, including the filtering of NAVI to 27 objects and the random sampling of reference-query pairs. The paper does not state whether all baselines are evaluated on the exact same reference-query pairs or whether the baseline numbers are taken from E2VG's paper. To avoid a favorable selection, please release the exact test pairs and verify that all methods are run under identical conditions, or use an independently constructed benchmark.
  4. [§4.3, Fig. 8] The refinement ablation compares the proposed two-side matching refinement with a 'naive refinement' baseline, but the naive baseline is not specified in sufficient detail. It is unclear whether the naive approach uses the same initialization, step count, and optimization variables as the proposed method, differing only in the two-sided generation. Without this control, the conclusion that two-side matching improves refinement accuracy at 15° could be confounded by implementation differences.
minor comments (6)
  1. [Abstract] The abstract contains a typo: 're leased' should be 'released'. Also, the URL appears with a space before the comma.
  2. [Table 1 caption] The caption has a typo: 'asd' should be 'and'.
  3. [Throughout] The dataset name is written inconsistently as 'NA VI' and 'NAVI'; please use a single spelling.
  4. [§3.3] The term 'score function' for f in Eq. (5) is nonstandard, since f is a loss to be minimized rather than a score in the statistical sense; consider renaming it 'matching loss' or 'score distillation loss' for clarity.
  5. [§4.2] The sentence 'the comparison between our method and them serves as an ablation study of the two-sided matching' is grammatically unclear; consider rephrasing.
  6. [§4.4] The phrase 'restricts the application in some real-time scenarios' should be 'restricts its application in some real-time scenarios'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the pose estimate is obtained by optimizing a pretrained diffusion model, and no step reduces the output to the input by construction.

full rationale

The paper's derivation chain is an optimization-based pose search: Eq. (3) defines a two-side image matching objective, Eq. (4) approximates it with the DreamFusion score-distillation loss, and Eq. (5) is the resulting Monte Carlo score function minimized over pose hypotheses. The relation between Eq. (3) and Eq. (4) is an explicit approximation ('we follow DreamFusion to approximate this problem'), not an identity, so the pose estimate is not encoded in the inputs by construction. The score function depends on the pretrained Zero123 denoiser as an oracle; its minimum is asserted to correspond to the true pose, but this is an empirical property rather than a circular reduction. The self-citation to the authors' prior E2VG appears only for the evaluation benchmark and preprocessing ('we follow [58] to transform Iq' and 'the same benchmark as E2VG'), which is a normal experimental practice and not load-bearing for the theoretical derivation. Any concerns about hyperparameter selection on the NAVI test set or the unvalidated proxy in Eq. (4) are correctness risks, not circularity. No step reduces, by definition or self-citation, to its own inputs.

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

The method relies on a pre-trained Zero123 diffusion model and on the DreamFusion score loss as a matching proxy. The key free parameters (N, M, t, refinement steps) are tuned on the NAVI test set, which raises the circularity burden. No new physical entities are introduced.

free parameters (4)
  • N (number of intermediate viewpoints) = 64
    Chosen from NAVI ablations (Table 3); higher N gives marginal gains.
  • M (Monte-Carlo noise samples) = 4
    Chosen from NAVI ablations (Table 4) as a balance between accuracy and runtime.
  • time step t = 0.4
    Chosen from NAVI ablations (Table 5); best accuracy at moderate noise.
  • refinement iterations = 3
    Chosen from NAVI ablation (Fig. 8); accuracy plateaus after 3.
assumptions (4)
  • domain assumption Zero123 provides reliable novel-view generation for small viewpoint changes
    The whole method relies on the pre-trained Zero123 model to generate intermediate views; Sec. 3.1 and Fig. 2.
  • ad hoc to paper The DreamFusion score distillation loss is a valid proxy for image matching distance in Eq. (3)
    Eq. (4) replaces the generated-image matching objective with the noise-prediction loss without proof; Sec. 3.3.
  • domain assumption The reference image pose (theta_r, phi_r) is known and the object coordinate system is canonical
    Sec. 3: 'we assume that the object is located at the origin... and the object pose of the reference image is known'.
  • domain assumption Object is centered and up direction aligned with gravity after preprocessing following [58]
    Sec. 3: 'We detect the object in Iq and follow [58] to transform Iq to look at the object'.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generalizable Single-view Object Pose Estimation by Two-side Generating and Matching." pith.science (2026). https://pith.science/paper/5MCI2RKP

@misc{pith2026241115860,
  author       = {Pith},
  title        = {Pith review of: Generalizable Single-view Object Pose Estimation by Two-side Generating and Matching},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5MCI2RKP}},
  note         = {Machine review of arXiv:2411.15860}
}
read the original abstract

In this paper, we present a novel generalizable object pose estimation method to determine the object pose using only one RGB image. Unlike traditional approaches that rely on instance-level object pose estimation and necessitate extensive training data, our method offers generalization to unseen objects without extensive training, operates with a single reference image of the object, and eliminates the need for 3D object models or multiple views of the object. These characteristics are achieved by utilizing a diffusion model to generate novel-view images and conducting a two-sided matching on these generated images. Quantitative experiments demonstrate the superiority of our method over existing pose estimation techniques across both synthetic and real-world datasets. Remarkably, our approach maintains strong performance even in scenarios with significant viewpoint changes, highlighting its robustness and versatility in challenging conditions. The code will be re leased at https://github.com/scy639/Gen2SM.

Figures

Figures reproduced from arXiv: 2411.15860 by the authors.

Figure 1
Figure 1. Our method excels at accurately estimating object poses [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. (a) Utilizing Zero123 [34] to directly generate an image from the viewpoint of the query image Iq based on the reference image Ir. Due to the significant viewpoint change, the generated image exhibits low quality, hindering accurate matching with the query image. (b) Instead, we leverage Zero123 [34] to generate images from intermediate viewpoints between the query image Iq and the reference image Ir. These generate… view at source ↗
Figure 3
Figure 3. The overview of our method. Given a reference image [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: We generate images G(z, Ir, ∆θri, ∆ϕri) of Ir and G(z, Iq, ∆θqi, ∆ϕqi) of Iq on N intermediate viewpoints {ϕi, θi|i = 1, ..., N}, which are sampled evenly from the upper hemisphere of the object. When the assumed ϕq, θq is correct, for each i, G(z, Iq, ∆θqi, ∆ϕqi) and …
Figure 5
Figure 5. Figure 5: As shown in Problem 4, we approximate the orig￾inal two-side matching problem by minimizing the loss func￾tion proposed by Poole et al. [49]. For a generated image set Gr = {Ir→i|i = 1, 2, ..., N} = {G(z, Ir, ∆θri, ∆ϕri)|i = 1, 2, ..., N}, we add noise to Ir→i to obtai…
Figure 6
Figure 6. Figure 6: Visual Comparison on testing sets [13, 22]. More results can be found in the supplementary materials [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: More examples of generated images on intermediate viewpoints, similar to Fig. [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Ablation studies on the refinement number and generat [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

79 extracted references · 52 canonical work pages

  1. [1]

    Map-free visual relocalization: Metric pose relative to a single image

    Eduardo Arnold, Jamie Wynn, Sara Vicente, Guillermo Garcia-Hernando, Aron Monszpart, Victor Prisacariu, Dani- yar Turmukhambetov, and Eric Brachmann. Map-free visual relocalization: Metric pose relative to a single image. In European Conference on Computer Vision, pages 690–708. Springer, 2022. 3, 6, 7

    work page 2022

  2. [2]

    Zs6d: Zero-shot 6d object pose estimation using vision transformers

    Philipp Ausserlechner, David Haberger, Stefan Thalhammer, Jean-Baptiste Weibel, and Markus Vincze. Zs6d: Zero-shot 6d object pose estimation using vision transformers. arXiv preprint arXiv:2309.11986, 2023. 3

    arXiv 2023

  3. [3]

    Surf: Speeded up robust features

    Herbert Bay, Tinne Tuytelaars, and Luc Van Gool. Surf: Speeded up robust features. In Computer Vision–ECCV 2006: 9th European Conference on Computer Vision, Graz, Austria, May 7-13, 2006. Proceedings, Part I 9, pages 404–

    2006

  4. [4]

    Zoedepth: Zero-shot trans- fer by combining relative and metric depth

    Shariq Farooq Bhat, Reiner Birkl, Diana Wofk, Peter Wonka, and Matthias M ¨uller. Zoedepth: Zero-shot trans- fer by combining relative and metric depth. arXiv preprint arXiv:2302.12288, 2023. 6, 7

    arXiv 2023

  5. [5]

    Ove6d: Ob- ject viewpoint encoding for depth-based 6d object pose es- timation

    Dingding Cai, Janne Heikkil ¨a, and Esa Rahtu. Ove6d: Ob- ject viewpoint encoding for depth-based 6d object pose es- timation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages 6803–6813, June 2022. 3

    work page 2022

  6. [6]

    Crt-6d: Fast 6d object pose estimation with cascaded refinement transformers

    Pedro Castro and Tae-Kyun Kim. Crt-6d: Fast 6d object pose estimation with cascaded refinement transformers. In Proceedings of the IEEE/CVF Winter Conference on Appli- cations of Computer Vision (WACV), pages 5746–5755, Jan- uary 2023. 3

    work page 2023

  7. [7]

    Posematcher: One-shot 6d object pose estimation by deep feature matching

    Pedro Castro and Tae-Kyun Kim. Posematcher: One-shot 6d object pose estimation by deep feature matching. In 2023 IEEE/CVF International Conference on Computer Vi- sion Workshops (ICCVW), pages 2140–2149, 2023. 2, 3

    work page 2023

  8. [8]

    Shapenet: An information-rich 3d model repository

    Angel X Chang, Thomas Funkhouser, Leonidas Guibas, Pat Hanrahan, Qixing Huang, Zimo Li, Silvio Savarese, Manolis Savva, Shuran Song, Hao Su, et al. Shapenet: An information-rich 3d model repository. arXiv preprint arXiv:1512.03012, 2015. 2, 3

    arXiv 2015

Show all 79 references
  1. [9]

    Id-pose: Sparse-view camera pose estimation by inverting diffusion models

    Weihao Cheng, Yan-Pei Cao, and Ying Shan. Id-pose: Sparse-view camera pose estimation by inverting diffusion models. arXiv preprint arXiv:2306.17140, 2023. 2, 3, 4, 6, 7

    2023 arXiv

  2. [10]

    Universal correspondence network

    Christopher B Choy, JunYoung Gwak, Silvio Savarese, and Manmohan Chandraker. Universal correspondence network. Advances in neural information processing systems , 29,

  3. [11]

    3dposelite: A compact 3d pose estimation using node em- beddings

    Meghal Dani, Karan Narain, and Ramya Hebbalaguppe. 3dposelite: A compact 3d pose estimation using node em- beddings. In 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), pages 1877–1886, 2021. 3

    2021

  4. [12]

    Objaverse: A universe of annotated 3d objects

    Matt Deitke, Dustin Schwenk, Jordi Salvador, Luca Weihs, Oscar Michel, Eli VanderBilt, Ludwig Schmidt, Kiana Ehsani, Aniruddha Kembhavi, and Ali Farhadi. Objaverse: A universe of annotated 3d objects. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Rec...

    2023

  5. [13]

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

    Laura Downs, Anthony Francis, Nate Koenig, Brandon Kin- man, Ryan Hickman, Krista Reymann, Thomas B McHugh, and Vincent Vanhoucke. Google scanned objects: A high- quality dataset of 3d scanned household items. In 2022 In- ternational Conference on Robotics and Automation (ICRA...

    2022

  6. [14]

    Measurement of areas on a sphere using fibonacci and latitude–longitude lattices

    ´Alvaro Gonz´alez. Measurement of areas on a sphere using fibonacci and latitude–longitude lattices. Mathematical geo- sciences, 42:49–64, 2010. 5

    2010

  7. [15]

    Zero-shot category-level object pose estimation

    Walter Goodwin, Sagar Vaze, Ioannis Havoutis, and Ingmar Posner. Zero-shot category-level object pose estimation. In European Conference on Computer Vision, pages 516–532. Springer, 2022. 3

    2022

  8. [16]

    Roth, and Vincent Lepetit

    Alexander Grabner, Yaming Wang, Peizhao Zhang, Peihong Guo, Tong Xiao, Peter Vajda, Peter M. Roth, and Vincent Lepetit. Geometric correspondence fields: Learned differ- entiable rendering for 3d pose refinement in the wild, 2020. 3

    2020

  9. [17]

    Pseudo flow consistency for self-supervised 6d object pose estimation

    Yang Hai, Rui Song, Jiaojiao Li, David Ferstl, and Yinlin Hu. Pseudo flow consistency for self-supervised 6d object pose estimation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 14075–14085, 2023. 3

    2023

  10. [18]

    Onepose++: Keypoint-free one- shot object pose estimation without CAD models

    Xingyi He, Jiaming Sun, Yuang Wang, Di Huang, Hujun Bao, and Xiaowei Zhou. Onepose++: Keypoint-free one- shot object pose estimation without CAD models. In Ad- vances in Neural Information Processing Systems , 2022. 2, 3

    2022

  11. [19]

    Denoising dif- fusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising dif- fusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020. 3, 4

    2020

  12. [20]

    Centersnap: Single-shot multi-object 3d shape reconstruction and categorical 6d pose and size estimation

    Muhammad Zubair Irshad, Thomas Kollar, Michael Laskey, Kevin Stone, and Zsolt Kira. Centersnap: Single-shot multi-object 3d shape reconstruction and categorical 6d pose and size estimation. In 2022 International Conference on Robotics and Automation (ICRA) , pages 10632–10640....

    2022

  13. [21]

    Shapo: Im- plicit representations for multi-object shape, appearance, and pose optimization

    Muhammad Zubair Irshad, Sergey Zakharov, Rares Ambrus, Thomas Kollar, Zsolt Kira, and Adrien Gaidon. Shapo: Im- plicit representations for multi-object shape, appearance, and pose optimization. In European Conference on Computer Vi- sion, pages 275–292. Springer, 2022. 3

    2022

  14. [22]

    NA VI: Category- agnostic image collections with high-quality 3d shape and pose annotations

    Varun Jampani, Kevis-Kokitsi Maninis, Andreas Engel- hardt, Arjun Karpur, Karen Truong, Kyle Sargent, Stefan Popov, Andre Araujo, Ricardo Martin-Brualla, Kaushal Pa- tel, Daniel Vlasic, Vittorio Ferrari, Ameesh Makadia, Ce Liu, Yuanzhen Li, and Howard Zhou. NA VI: Category- ag...

    2023

  15. [23]

    Cotr: Correspondence transformer for matching across images

    Wei Jiang, Eduard Trulls, Jan Hosang, Andrea Tagliasacchi, and Kwang Moo Yi. Cotr: Correspondence transformer for matching across images. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 6207– 6217, 2021. 3

    2021

  16. [24]

    End-to-end recovery of human shape and pose

    Angjoo Kanazawa, Michael J Black, David W Jacobs, and Jitendra Malik. End-to-end recovery of human shape and pose. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7122–7131, 2018. 3

    2018

  17. [25]

    Learning 3d human dynamics from video

    Angjoo Kanazawa, Jason Y Zhang, Panna Felsen, and Jiten- dra Malik. Learning 3d human dynamics from video. In Proceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 5614–5623, 2019. 1

    2019

  18. [26]

    Vibe: Video inference for human body pose and shape estimation

    Muhammed Kocabas, Nikos Athanasiou, and Michael J Black. Vibe: Video inference for human body pose and shape estimation. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 5253–5263, 2020. 1

    2020

  19. [27]

    Nerf- pose: A first-reconstruct-then-regress approach for weakly- supervised 6d object pose estimation

    Fu Li, Shishir Reddy Vutukur, Hao Yu, Ivan Shugurov, Benjamin Busam, Shaowu Yang, and Slobodan Ilic. Nerf- pose: A first-reconstruct-then-regress approach for weakly- supervised 6d object pose estimation. In 2023 IEEE/CVF In- ternational Conference on Computer Vision Workshops...

    2023

  20. [28]

    Sd-pose: Structural discrepancy aware category-level 6d object pose estimation

    Guowei Li, Dongchen Zhu, Guanghui Zhang, Wenjun Shi, Tianyu Zhang, Xiaolin Zhang, and Jiamao Li. Sd-pose: Structural discrepancy aware category-level 6d object pose estimation. In Proceedings of the IEEE/CVF Winter Con- ference on Applications of Computer Vision (WACV), pages ...

    2023

  21. [29]

    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. In European Conference Computer Vision (ECCV), 2018. 2, 3

    2018

  22. [30]

    Relpose++: Recovering 6d poses from sparse-view observations

    Amy Lin, Jason Y Zhang, Deva Ramanan, and Shubham Tul- siani. Relpose++: Recovering 6d poses from sparse-view observations. arXiv preprint arXiv:2305.04926, 2023. 3, 4, 6, 7

    2023 arXiv

  23. [31]

    Learning to estimate object poses without real image anno- tations

    Haotong Lin, Sida Peng, Zhize Zhou, and Xiaowei Zhou. Learning to estimate object poses without real image anno- tations. In IJCAI, pages 1159–1165, 2022. 3

    2022

  24. [32]

    Sift flow: Dense correspondence across scenes and its applications

    Ce Liu, Jenny Yuen, and Antonio Torralba. Sift flow: Dense correspondence across scenes and its applications. IEEE transactions on pattern analysis and machine intelligence , 33(5):978–994, 2010. 1, 3

    2010

  25. [33]

    One-2-3-45: Any single image to 3d mesh in 45 seconds without per-shape optimization.arXiv preprint arXiv:2306.16928, 2023

    Minghua Liu, Chao Xu, Haian Jin, Linghao Chen, Zexiang Xu, Hao Su, et al. One-2-3-45: Any single image to 3d mesh in 45 seconds without per-shape optimization.arXiv preprint arXiv:2306.16928, 2023. 6

    2023 arXiv

  26. [34]

    Zero-1-to-3: Zero-shot one image to 3d object

    Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tok- makov, Sergey Zakharov, and Carl V ondrick. Zero-1-to-3: Zero-shot one image to 3d object. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 9298–9309, 2023. 2, 3, 4, 6

    2023

  27. [35]

    Syncdreamer: Gen- erating multiview-consistent images from a single-view im- age

    Yuan Liu, Cheng Lin, Zijiao Zeng, Xiaoxiao Long, Lingjie Liu, Taku Komura, and Wenping Wang. Syncdreamer: Gen- erating multiview-consistent images from a single-view im- age. arXiv preprint arXiv:2309.03453, 2023. 3, 9

    2023 arXiv

  28. [36]

    Gen6d: Generalizable model-free 6-dof object pose estimation from rgb images

    Yuan Liu, Yilin Wen, Sida Peng, Cheng Lin, Xiaoxiao Long, Taku Komura, and Wenping Wang. Gen6d: Generalizable model-free 6-dof object pose estimation from rgb images. In European Conference on Computer Vision, pages 298–315. Springer, 2022. 2, 3, 4

    2022

  29. [37]

    Wonder3d: Sin- gle image to 3d using cross-domain diffusion.arXiv preprint arXiv:2310.15008, 2023

    Xiaoxiao Long, Yuan-Chen Guo, Cheng Lin, Yuan Liu, Zhiyang Dou, Lingjie Liu, Yuexin Ma, Song-Hai Zhang, Marc Habermann, Christian Theobalt, et al. Wonder3d: Sin- gle image to 3d using cross-domain diffusion.arXiv preprint arXiv:2310.15008, 2023. 3, 9

    arXiv

  30. [38]

    Distinctive image features from scale- invariant keypoints

    David G Lowe. Distinctive image features from scale- invariant keypoints. International journal of computer vi- sion, 60:91–110, 2004. 3, 6, 7

    2004

  31. [39]

    Virtual correspondence: Hu- mans as a cue for extreme-view geometry

    Wei-Chiu Ma, Anqi Joyce Yang, Shenlong Wang, Raquel Urtasun, and Antonio Torralba. Virtual correspondence: Hu- mans as a cue for extreme-view geometry. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 15924–15934, 2022. 1

    2022

  32. [40]

    Vnect: Real-time 3d human pose estimation with a single rgb cam- era

    Dushyant Mehta, Srinath Sridhar, Oleksandr Sotnychenko, Helge Rhodin, Mohammad Shafiei, Hans-Peter Seidel, Weipeng Xu, Dan Casas, and Christian Theobalt. Vnect: Real-time 3d human pose estimation with a single rgb cam- era. Acm transactions on graphics (tog) , 36(4):1–14, 2017. 3

    2017

  33. [41]

    Relative camera pose estimation using convolutional neural networks

    Iaroslav Melekhov, Juha Ylioinas, Juho Kannala, and Esa Rahtu. Relative camera pose estimation using convolutional neural networks. In Advanced Concepts for Intelligent Vi- sion Systems: 18th International Conference, ACIVS 2017, Antwerp, Belgium, September 18-21, 2017, Procee...

    2017

  34. [42]

    T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models

    Chong Mou, Xintao Wang, Liangbin Xie, Jian Zhang, Zhon- gang Qi, Ying Shan, and Xiaohu Qie. T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models. arXiv preprint arXiv:2302.08453, 2023. 3

    2023 arXiv

  35. [43]

    Nope: Novel ob- ject pose estimation from a single image

    Van Nguyen Nguyen, Thibault Groueix, Yinlin Hu, Math- ieu Salzmann, and Vincent Lepetit. Nope: Novel ob- ject pose estimation from a single image. arXiv preprint arXiv:2303.13612, 2023. 3

    2023 arXiv

  36. [44]

    Gigapose: Fast and robust novel ob- ject pose estimation via one correspondence

    Van Nguyen Nguyen, Thibault Groueix, Mathieu Salzmann, and Vincent Lepetit. Gigapose: Fast and robust novel ob- ject pose estimation via one correspondence. arXiv preprint arXiv:2311.14155, 2023. 3

    2023 arXiv

  37. [45]

    Templates for 3d object pose es- timation revisited: Generalization to new objects and robust- ness to occlusions

    Van Nguyen Nguyen, Yinlin Hu, Yang Xiao, Mathieu Salz- mann, and Vincent Lepetit. Templates for 3d object pose es- timation revisited: Generalization to new objects and robust- ness to occlusions. In Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2022. 3

    2022

  38. [46]

    Zephyr: Zero-shot pose hypothesis rating

    Brian Okorn, Qiao Gu, Martial Hebert, and David Held. Zephyr: Zero-shot pose hypothesis rating. In 2021 IEEE In- ternational Conference on Robotics and Automation (ICRA), pages 14141–14148. IEEE, 2021. 3

    2021

  39. [47]

    Latentfusion: End-to-end differentiable reconstruction and rendering for unseen object pose estimation

    Keunhong Park, Arsalan Mousavian, Yu Xiang, and Dieter Fox. Latentfusion: End-to-end differentiable reconstruction and rendering for unseen object pose estimation. InProceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition, 2020. 3

    2020

  40. [48]

    Cornet: Generic 3d corners for 6d pose estimation of new objects without retraining

    Giorgia Pitteri, Slobodan Ilic, and Vincent Lepetit. Cornet: Generic 3d corners for 6d pose estimation of new objects without retraining. In 2019 IEEE/CVF International Confer- ence on Computer Vision Workshop (ICCVW), pages 2807– 2815, 2019. 3

    2019

  41. [49]

    Dreamfusion: Text-to-3d using 2d diffusion

    Ben Poole, Ajay Jain, Jonathan T Barron, and Ben Milden- hall. Dreamfusion: Text-to-3d using 2d diffusion. arXiv preprint arXiv:2209.14988, 2022. 4, 5

    2022 arXiv

  42. [50]

    Com- mon objects in 3d: Large-scale learning and evaluation of real-life 3d category reconstruction

    Jeremy Reizenstein, Roman Shapovalov, Philipp Henzler, Luca Sbordone, Patrick Labatut, and David Novotny. Com- mon objects in 3d: Large-scale learning and evaluation of real-life 3d category reconstruction. In Proceedings of the IEEE/CVF International Conference on Computer Vi...

    2021

  43. [51]

    The 8- point algorithm as an inductive bias for relative pose predic- tion by vits

    Chris Rockwell, Justin Johnson, and David F Fouhey. The 8- point algorithm as an inductive bias for relative pose predic- tion by vits. In 2022 International Conference on 3D Vision (3DV), pages 1–11. IEEE, 2022. 3

    2022

  44. [52]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 3

    2022

  45. [53]

    Superglue: Learning feature matching with graph neural networks

    Paul-Edouard Sarlin, Daniel DeTone, Tomasz Malisiewicz, and Andrew Rabinovich. Superglue: Learning feature matching with graph neural networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4938–4947, 2020. 1, 3

    2020

  46. [54]

    Ocskb: An object com- ponent sketch knowledge base for fast 6d pose estimation

    Guangming Shi, Xuyang Li, Xuemei Xie, Mingxuan Yu, Chengwei Rao, and Jiakai Luo. Ocskb: An object com- ponent sketch knowledge base for fast 6d pose estimation. In Proceedings of the 31st ACM International Conference on Multimedia, MM ’23, page 5819–5827, New York, NY , USA, 2...

    2023

  47. [55]

    Osop: A multi-stage one shot object pose estimation frame- work

    Ivan Shugurov, Fu Li, Benjamin Busam, and Slobodan Ilic. Osop: A multi-stage one shot object pose estimation frame- work. In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 6825–6834, 2022. 3

    2022

  48. [56]

    Loftr: Detector-free local feature matching with transformers

    Jiaming Sun, Zehong Shen, Yuang Wang, Hujun Bao, and Xiaowei Zhou. Loftr: Detector-free local feature matching with transformers. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 8922–8931, 2021. 1, 3, 6, 7

    2021

  49. [57]

    Onepose: One-shot object pose estimation without cad mod- els

    Jiaming Sun, Zihao Wang, Siyu Zhang, Xingyi He, Hongcheng Zhao, Guofeng Zhang, and Xiaowei Zhou. Onepose: One-shot object pose estimation without cad mod- els. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 6825–6834,

  50. [58]

    Extreme Two-View Geometry From Object Poses with Diffusion Models

    Yujing Sun, Caiyi Sun, Yuan Liu, Yuexin Ma, and Siu Ming Yiu. Extreme Two-View Geometry From Object Poses with Diffusion Models. arXiv e-prints, page arXiv:2402.02800, Feb. 2024. 2, 3, 4, 6, 7

    2024 arXiv

  51. [59]

    Mvdiffusion++: A dense high-resolution multi-view diffusion model for sin- gle or sparse-view 3d object reconstruction

    Shitao Tang, Jiacheng Chen, Dilin Wang, Chengzhou Tang, Fuyang Zhang, Yuchen Fan, Vikas Chandra, Yasu- taka Furukawa, and Rakesh Ranjan. Mvdiffusion++: A dense high-resolution multi-view diffusion model for sin- gle or sparse-view 3d object reconstruction. arXiv preprint arXiv...

    2024 arXiv

  52. [60]

    Daisy: An effi- cient dense descriptor applied to wide-baseline stereo

    Engin Tola, Vincent Lepetit, and Pascal Fua. Daisy: An effi- cient dense descriptor applied to wide-baseline stereo. IEEE transactions on pattern analysis and machine intelligence , 32(5):815–830, 2009. 3

    2009

  53. [61]

    Metapose: Fast 3d pose from multiple views without 3d supervision

    Ben Usman, Andrea Tagliasacchi, Kate Saenko, and Avneesh Sud. Metapose: Fast 3d pose from multiple views without 3d supervision. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 6759–6770, 2022. 1, 3

    2022

  54. [62]

    Sv3d: Novel multi-view syn- thesis and 3d generation from a single image using latent video diffusion

    Vikram V oleti, Chun-Han Yao, Mark Boss, Adam Letts, David Pankratz, Dmitry Tochilkin, Christian Laforte, Robin Rombach, and Varun Jampani. Sv3d: Novel multi-view syn- thesis and 3d generation from a single image using latent video diffusion. arXiv preprint arXiv:2403.12008, 2024. 8

    2024 arXiv

  55. [63]

    Dtf-net: Category-level pose estima- tion and shape reconstruction via deformable template field

    Haowen Wang, Zhipeng Fan, Zhen Zhao, Zhengping Che, Zhiyuan Xu, Dong Liu, Feifei Feng, Yakun Huang, Xiuquan Qiao, and Jian Tang. Dtf-net: Category-level pose estima- tion and shape reconstruction via deformable template field. In Proceedings of the 31st ACM International Confe...

    2023

  56. [64]

    Pf-lrm: Pose-free large reconstruction model for joint pose and shape prediction

    Peng Wang, Hao Tan, Sai Bi, Yinghao Xu, Fujun Luan, Kalyan Sunkavalli, Wenping Wang, Zexiang Xu, and Kai Zhang. Pf-lrm: Pose-free large reconstruction model for joint pose and shape prediction. arXiv preprint arXiv:2311.12024, 2023. 3

    2023 arXiv

  57. [65]

    Dust3r: Geometric 3d vi- sion made easy

    Shuzhe Wang, Vincent Leroy, Yohann Cabon, Boris Chidlovskii, and Jerome Revaud. Dust3r: Geometric 3d vi- sion made easy. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 20697– 20709, 2024. 3

    2024

  58. [66]

    Object pose estimation via the aggregation of diffusion features

    Tianfu Wang, Guosheng Hu, and Hongguang Wang. Object pose estimation via the aggregation of diffusion features. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 10238–10247, 2024. 3

    2024

  59. [67]

    Bundletrack: 6d pose track- ing for novel objects without instance or category-level 3d models

    B Wen and Kostas E Bekris. Bundletrack: 6d pose track- ing for novel objects without instance or category-level 3d models. In IEEE/RSJ International Conference on Intelli- gent Robots and Systems, 2021. 3

    2021

  60. [68]

    BundleSDF: Neural 6-DoF tracking and 3D reconstruction of unknown objects

    Bowen Wen, Jonathan Tremblay, Valts Blukis, Stephen Tyree, Thomas M ¨uller, Alex Evans, Dieter Fox, Jan Kautz, and Stan Birchfield. BundleSDF: Neural 6-DoF tracking and 3D reconstruction of unknown objects. In CVPR, 2023. 3

    2023

  61. [69]

    Gccn: Geometric constraint co-attention network for 6d object pose estimation

    Yongming Wen, Yiquan Fang, Junhao Cai, Kimwa Tung, and Hui Cheng. Gccn: Geometric constraint co-attention network for 6d object pose estimation. In Proceedings of the 29th ACM International Conference on Multimedia, MM ’21, page 2671–2679, New York, NY , USA, 2021. Associa- ti...

    2021

  62. [70]

    Disp6d: Disentangled implicit shape and pose learning for scalable 6d pose esti- mation

    Yilin Wen, Xiangyu Li, Hao Pan, Lei Yang, Zheng Wang, Taku Komura, and Wenping Wang. Disp6d: Disentangled implicit shape and pose learning for scalable 6d pose esti- mation. In European Conference on Computer Vision, pages 404–421. Springer, 2022. 1

    2022

  63. [71]

    Pose from shape: Deep pose es- timation for arbitrary 3D objects

    Yang Xiao, Xuchong Qiu, Pierre-Alain Langlois, Mathieu Aubry, and Renaud Marlet. Pose from shape: Deep pose es- timation for arbitrary 3D objects. In British Machine Vision Conference (BMVC), 2019. 3

    2019

  64. [72]

    Barron, Alberto Rodriguez, Phillip Isola, and Tsung-Yi Lin

    Lin Yen-Chen, Pete Florence, Jonathan T. Barron, Alberto Rodriguez, Phillip Isola, and Tsung-Yi Lin. Inerf: Inverting neural radiance fields for pose estimation, 2021. 3

    2021

  65. [73]

    Fs6d: Few-shot 6d pose estimation of novel ob- jects

    He Yisheng, Wang Yao, Fan Haoqiang, Chen Qifeng, and Sun Jian. Fs6d: Few-shot 6d pose estimation of novel ob- jects. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2022. 3

    2022

  66. [74]

    Cameras as rays: Pose estimation via ray diffusion

    Jason Y Zhang, Amy Lin, Moneish Kumar, Tzu-Hsuan Yang, Deva Ramanan, and Shubham Tulsiani. Cameras as rays: Pose estimation via ray diffusion. arXiv preprint arXiv:2402.14817, 2024. 3

    2024 arXiv

  67. [75]

    Rel- pose: Predicting probabilistic relative rotation for single ob- jects in the wild

    Jason Y Zhang, Deva Ramanan, and Shubham Tulsiani. Rel- pose: Predicting probabilistic relative rotation for single ob- jects in the wild. In European Conference on Computer Vi- sion, pages 592–611. Springer, 2022. 3, 4, 6

    2022

  68. [76]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3836–3847, 2023. 3

    2023

  69. [77]

    Fusing local similarities for retrieval-based 3d orientation estimation of unseen objects, 2022

    Chen Zhao, Yinlin Hu, and Mathieu Salzmann. Fusing local similarities for retrieval-based 3d orientation estimation of unseen objects, 2022. 2, 3

    2022

  70. [78]

    Locposenet: Robust location prior for unseen object pose estimation,

    Chen Zhao, Yinlin Hu, and Mathieu Salzmann. Locposenet: Robust location prior for unseen object pose estimation,

  71. [79]

    3d-aware hypothesis & verification for generalizable relative object pose estimation

    Chen Zhao, Tong Zhang, and Mathieu Salzmann. 3d-aware hypothesis & verification for generalizable relative object pose estimation. Proceedings of the International Confer- ence on Learning Representations, 2024. 6, 7

    2024

Pith tools

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