Pragmatist: Multiview Conditional Diffusion Models for High-Fidelity 3D Reconstruction from Unposed Sparse Views

Chunhui Zhao; Songchun Zhang

REVIEW 4 major objections 6 minor 55 references

Pragmatist: Multiview Conditional Diffusion Models for High-Fidelity 3D Reconstruction from Unposed Sparse Views

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Generate first, reconstruct later: 3D from unposed views

desk verdict A well-engineered pipeline that may be a real step forward for unposed sparse-view reconstruction, but the reported evidence does not yet prove the headline gains. read the letter →

arxiv 2412.08412 v2 pith:4HVO2MJG submitted 2024-12-11 cs.CV

Songchun Zhang , Chunhui Zhao This is my paper · ORCID

classification cs.CV

keywords 3Dreconstructionunposedsparseviewsmulti-viewdiffusionnovelviewsynthesiscanonicalcamerasystemtriplanerepresentationposerefinement

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper argues that sparse unposed 3D reconstruction should be split into two stages: first, a pose-free multi-view diffusion model generates a complete set of canonical views of the object; second, geometry is reconstructed from those generated views. The claim is that this reformulation avoids the direct, ill-posed regression from pixels to 3D and lets the model plausibly fill in unseen regions using learned geometric priors. If true, casual sparse captures—images with no camera poses—can yield high-fidelity meshes and textures, and even single images become competitive with dedicated single-view methods. Recovering input-view poses by inverting the reconstructed 3D model, rather than estimating them from 2D images, is part of the mechanism that makes this work.

What carries the argument

The central object is the canonical camera system: a fixed set of target views at 0° elevation and distance 1.5 with azimuth offsets β, β±45°, β±90°, β±135°, and β±180°. The diffusion model conditions on unposed inputs through a lightweight encoder and injects pixel-level camera conditions via Plücker ray embeddings, while extending self-attention to cross-frame attention across target views and cross-attention between input and target views. This machinery embodies the pose-free consistency learning that maps arbitrary casual viewpoints into a shared canonical frame. The triplane representation produced by the reconstruction model is the object that gets inverted for pose recovery and refined with input-view textures.

What would settle it

Render test objects from extreme elevations and distances far from the canonical 0° elevation / 1.5 distance setting, run the pipeline, and measure the multi-view consistency of the generated canonical views (e.g., feature-metric reprojection error under recovered poses) and the final mesh Chamfer distance. If consistency or reconstruction degrades sharply at out-of-distribution viewpoints, the canonical camera system is the load-bearing constraint.

Watch

Extended reading notes

Core claim

Pragmatist's central claim is that unposed sparse-view 3D reconstruction can be recast as conditional novel view synthesis in a canonical object coordinate system. A multi-view conditional diffusion model, conditioned on arbitrary unposed input views via a lightweight encoder and cross-frame attention, generates several novel views at fixed elevation and distance with azimuth offsets. These canonical views are fed to a feed-forward triplane reconstruction model, which predicts a NeRF and then a mesh via differentiable marching cubes and rasterization. Input-view poses are recovered by inverting the triplane representation through differentiable rendering, and texture is refined using the high-resolution input views. The paper's experiments support that generating complete observations before reconstruction improves fidelity over direct pose-free prediction and over pipelines that estimate poses in image space.

Load-bearing premise

The multi-view diffusion model can learn, without any input camera poses, to map arbitrary unposed views into a shared canonical object coordinate system and produce geometrically consistent novel views from them.

Editorial extensions

If this is right

Sparse unposed inputs can be turned into fully observed reconstruction problems, so the reconstruction stage sees complete observations rather than partial ones.
Recovering input-view poses by inverting the generated 3D representation is far more accurate than direct image-space pose estimation, with rotation error of 2.182 versus 67.580 on the GSO dataset.
The pipeline supports flexible numbers of input views, including a single image, and its reconstruction quality improves as more unposed views are added.
Separating density and color MLPs lets appearance refinement improve texture fidelity without degrading geometry.

Reading between the lines

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

The generate-then-reconstruct paradigm suggests that any sufficiently strong multi-view generative prior could replace the diffusion model, and the canonical coordinate design could be relaxed to estimate elevation and scale per object, extending the method to in-the-wild captures with varying camera intrinsics.
Because the generator is trained without pose conditioning on inputs, the same architecture could be trained on large-scale video data, potentially lifting the approach from objects to scenes—a direction the paper states as future work.
A testable extension is to provide the diffusion model with noisy or partial pose priors, such as rough azimuth ordering, and measure whether consistency and reconstruction improve; if they do, the pose-free encoding is not fully capturing viewpoint information.

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.

Desk Editor's Note

read the letter

Short version: a plausible, well-engineered pipeline that may be a real step forward for unposed sparse-view reconstruction, but as reported the evidence does not yet support the headline numbers.

The genuinely new element is the framing: instead of predicting geometry directly from unposed inputs, generate a complete set of canonical views with a pose-free multiview diffusion model, then reconstruct from those views and recover input poses by inverting the triplane. The individual pieces—multi-view diffusion, LRM, DiffMC, pose/refinement—are borrowed, but the integrated pose-free canonicalization is not present verbatim in the cited work. The ablations are in the right direction and do support the claim that generative priors and the refinement stage matter. The method also handles a variable number of input views, which is practically useful.

The soft spots are about evidence, not the core idea. There is no code, no data, no error bars, an unspecified Objaverse filter, an unclear evaluation subset, and no explanation of what 'Ours Pred.' means in Table 2 (as opposed to 'Ours' with unposed inputs). The pose benchmark in Table 3 is self-referential: errors are measured relative to the model's own canonical mesh, not ground-truth object coordinates, so it cannot establish that the canonicalization is actually correct. There is also an expository inconsistency in Eq. 1, which lists Pcond as conditioning even though the text elsewhere says poses of condition views are not used. Those are fixable, but they matter because canonicalization is the load-bearing step. If the canonical frame drifts, Eq. 4 and Eq. 5 inherit the error. The paper does not currently show that this doesn't happen.

Overall: this is a serious engineering effort on a practical problem. The reader should not accept the central benchmark claim without artifacts and a clean pose-evaluation protocol, but the idea deserves referee time. I'd send it to review with a request for code/data, error bars, clarification of the evaluation protocol, and a pose benchmark against ground truth rather than the model's own frame.

Referee Report

4 major / 6 minor

Summary. The paper proposes Pragmatist, a three-stage pipeline for reconstructing a 3D object from sparse unposed views. First, a pose-free multi-view conditional diffusion model generates several novel views in a canonical object coordinate system. Second, a feed-forward large reconstruction model predicts a triplane NeRF and extracts a mesh from these generated views. Third, the input views are registered to the canonical frame by optimizing camera poses against the reconstructed mesh, and the texture is refined using the high-resolution input views. The authors report improvements over pose-free and pose-based baselines on GSO and OmniObject3D, plus ablation studies supporting the role of the generative prior and refinement.

Significance. If the results hold, the paper offers a practical reformulation of unposed sparse-view reconstruction: instead of solving an ill-posed inverse problem directly, the method synthesizes canonical observations with a generative model and then uses a feed-forward reconstructor plus test-time refinement. This is a plausible and potentially useful system contribution, and the empirical comparison covers a broad set of recent baselines. The paper does not provide code or pre-trained models, and the quantitative evidence currently lacks statistical robustness and a non-self-referential pose evaluation.

major comments (4)

[Table 3, Eq. (4)] The pose accuracy evaluation in Table 3 is self-referential and needs clarification. The text describes pose recovery as optimizing the camera pose p of I_cond against the mesh obtained from the feed-forward reconstruction (Eq. 4), but it does not specify what the ground-truth pose is for unposed input views. The CD/IoU evaluation aligns meshes with ICP, but no analogous alignment step is described for pose evaluation. If the reported rotation and translation errors are computed with respect to the model's own canonical mesh, then Table 3 measures only the convergence of the pose optimization, not the accuracy of the canonicalization. This is load-bearing because the refinement in Eq. (5) and the 'Ours Pred.' row in Table 2 depend on these poses. Please specify the reference frame, how ground-truth poses are obtained, and whether any alignment to the GT object is applied before computing R. error and T. error.
[Eq. (1) and 'Conditioning Feature Extraction'] There is an inconsistency between the formal model in Eq. (1), which conditions generation on Icond and Pcond, and the later statement that the model works 'without explicitly using the camera pose information of the conditional views.' The conditioning feature extraction paragraph describes only an image encoder; it does not explain how Pcond is encoded or whether it is dropped at inference. If Pcond is used only during training to define the canonical target views relative to the condition views, that should be stated explicitly. As written, the reader cannot determine whether the model at inference is being asked to do something different from its training objective, which is central to the canonicalization claim.
[Tables 2-5 and 'Datasets' paragraph] The evaluation protocol is under-specified. No error bars, number of test objects, number of sampled views per object, or random seed information is reported, so it is unclear whether the improvements in Tables 2, 4, and 5 are statistically significant. Furthermore, the rows 'Ours' and 'Ours Pred.' in Table 2 are not defined in the text; the reader cannot tell what each variant does (e.g., whether 'Ours' uses the canonical generated views without refinement, or uses predicted poses for the input views). Please provide a precise evaluation protocol, standard deviations, and a clear definition of every reported variant.
[Canonical Camera System, Eq. (5)] The load-bearing assumption that the diffusion model can map arbitrary unposed views to a canonical object coordinate system is not quantitatively validated. The 'Canonical Camera System' fixes target views at elevation 0 and distance 1.5 with azimuth offsets, but the condition views have arbitrary pose, and the model must infer the object's orientation, scale, and center from RGB alone. If the inferred canonical frame drifts, the pose optimization in Eq. (4) and the texture refinement in Eq. (5) are anchored to a wrong coordinate system, so the photometric loss can blur or warp instead of sharpen. The only quantitative evidence of canonicalization is Table 3, which is self-referential as noted above. Please evaluate canonicalization directly, for example by comparing the generated canonical views to ground-truth canonical renderings of the ICP-aligned GT mesh, or by reporting failure cases.

minor comments (6)

[Table 4] In the 'Ours' row of Table 4, the pose column contains 'Pred.' at the end of the row, which appears to be a typo.
[Eqs. (2)-(3) and Implementation details] The loss weight λ in Eq. (2) is defined only in the experimental section, and the text does not clearly connect λ in Eq. (2) to λp in the implementation. Please give the value in the equation or refer to the implementation paragraph at the first use.
[Table 1] The abbreviation 'Gen Occ.' in Table 1 is not expanded in the table or the text; please spell out 'Generative Occlusion' or whatever is intended.
[References] References Wei et al. 2023a and 2023b appear to be the same paper (NeuralManifold) and should be merged; the supplementary material cites Sohl-Dickstein without a corresponding reference entry.
[Ablation of Virtual Camera Guided Refinement] The claim that 'optimizing the triplane feature and color MLP at the same time achieves better results than optimizing the color MLP alone' is not supported by any table or figure; please provide quantitative evidence.
[Pose Estimation Results] The comparison with HLoc in Table 3 is unclear: HLoc is a full SfM pipeline, and it is not stated whether it was run under the same sparse-view setting as the other methods. Please clarify the exact input to each baseline.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the pipeline's stages are externally trained and benchmarked, and the flagged pose-metric concern is an evaluation-detail issue, not a demonstrated self-referential reduction.

full rationale

The paper's derivation chain is not circular. The central claim is that generating canonical multi-view observations with a pose-free conditional diffusion model before feed-forward triplane reconstruction improves sparse unposed 3D reconstruction. Each stage is grounded externally: the multiview generator and the reconstruction model are trained on filtered Objaverse renderings, and the headline tables (Tables 2 and 4) report novel-view and mesh metrics against held-out GSO/OmniObject3D ground truth with an explicitly described GT-alignment/ICP procedure for geometry. Eq. (4) is a standard test-time camera registration against the reconstructed model; it is not a renamed input, and the paper does not state that the Table 3 pose errors are measured with respect to the model's own canonical mesh. Eq. (5) is legitimate test-time texture fitting to the input views, not a prediction forced by construction. The only self-citation (Zhang and Zhao 2023) appears in a related-work list and is not load-bearing. Although the pose-evaluation protocol in Table 3 is under-specified (no explicit GT canonical-frame alignment description), absence of that detail is an evaluation-transparency concern, not a demonstrated circular reduction.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claim rests on design choices baked into the pipeline (canonical camera geometry, loss weights, data filtering) and domain assumptions about generalization (Objaverse-to-GSO transfer, pose-free canonicalization, segmentation availability). These are not derived or validated with sensitivity analysis; if any of them is wrong for a particular input, the whole cascade can fail.

free parameters (4)

Canonical camera distance and elevation = 1.5, 0 degrees
Chosen by hand to remove scale and elevation ambiguity; the diffusion model must map arbitrary input views into this fixed frame.
Canonical azimuth set = beta, beta +/- 45, +/- 90, +/- 135, +/- 180 degrees
Hand-selected set of generated views; reconstruction quality depends on this choice.
Loss weights in Eqs (2) and (3) = lambda_p=0.5, lambda_d=0.5, lambda_m=1, lambda_n=1
Manually chosen hyperparameters with no sensitivity analysis reported.
Objaverse quality filter = about 240k assets
The phrase 'quality assets' is an undefined ad hoc criterion for training data selection.

assumptions (5)

domain assumption A multi-view latent diffusion model with cross-frame attention can learn 3D-consistent canonical views from unposed sparse inputs without explicit pose or intrinsics.
This is the core assumption of the multiview conditional generator; the paper offers qualitative examples but no failure analysis or formal justification.
domain assumption Training on filtered Objaverse renderings transfers to GSO, OmniObject3D, and in-the-wild or text-to-image images.
The datasets section states the filter but gives no domain-gap analysis or per-category breakdown.
domain assumption Foreground segmentation masks are available to center objects before pose optimization.
The method says 'we use foreground segmentation to obtain the object mask' but does not specify the segmentation model or how failures affect pose accuracy.
standard math Volume rendering, DiffMC, and differentiable rasterization are correctly implemented and standard.
These are established tools; no proof is expected, but errors in these components would propagate into all results.
ad hoc to paper The simplified spherical camera model p={alpha, beta, gamma, r} captures the input view geometry after homography centering.
The search space is reduced to four parameters, which assumes the object-centered homography removes other degrees of freedom; this is a bespoke design choice.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Pragmatist: Multiview Conditional Diffusion Models for High-Fidelity 3D Reconstruction from Unposed Sparse Views." pith.science (2026). https://pith.science/paper/4HVO2MJG

@misc{pith2026241208412,
  author       = {Pith},
  title        = {Pith review of: Pragmatist: Multiview Conditional Diffusion Models for High-Fidelity 3D Reconstruction from Unposed Sparse Views},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4HVO2MJG}},
  note         = {Machine review of arXiv:2412.08412}
}

read the original abstract

Inferring 3D structures from sparse, unposed observations is challenging due to its unconstrained nature. Recent methods propose to predict implicit representations directly from unposed inputs in a data-driven manner, achieving promising results. However, these methods do not utilize geometric priors and cannot hallucinate the appearance of unseen regions, thus making it challenging to reconstruct fine geometric and textural details. To tackle this challenge, our key idea is to reformulate this ill-posed problem as conditional novel view synthesis, aiming to generate complete observations from limited input views to facilitate reconstruction. With complete observations, the poses of the input views can be easily recovered and further used to optimize the reconstructed object. To this end, we propose a novel pipeline Pragmatist. First, we generate a complete observation of the object via a multiview conditional diffusion model. Then, we use a feed-forward large reconstruction model to obtain the reconstructed mesh. To further improve the reconstruction quality, we recover the poses of input views by inverting the obtained 3D representations and further optimize the texture using detailed input views. Unlike previous approaches, our pipeline improves reconstruction by efficiently leveraging unposed inputs and generative priors, circumventing the direct resolution of highly ill-posed problems. Extensive experiments show that our approach achieves promising performance in several benchmarks.

Figures

Figures reproduced from arXiv: 2412.08412 by the authors.

**Figure 2.** Overview of our pipeline. Our core insight is to generate additional novel views in the canonical object coordinate by conditioning with casual viewpoints to solve sparse unposed view reconstruction. The multi-view conditional generator uses a lightweight encoder to extract input view features and implicitly models the 3D consistency between the input view and the output view in the canonical coordinate system via c… view at source ↗

**Figure 3.** Qualitative comparison of unposed sparse-view 3D reconstruction results on the GSO dataset. Our method can pro [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Qualitative comparison of single view 3D re [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Ablation Studies. The top image illustrates the effectiveness of our reconstruction model, the bottom left image shows the effect of our refinement step, and the bottom right image demonstrates the effect of generative priors. Method Pose GSO OmniObject3D CD↓ IoU↑ CD…

**Figure 6.** Figure 6: Qualitative comparison of single view 3D reconstruction results on the GSO and in-the-wild datasets. Though our [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]

**Figure 7.** Figure 7: Ablation Study on Generative Priors. As shown above, incorporating generative priors enables the synthesis of plausible textures in unseen regions of objects. of all views. The forward process of the multiview diffusion model is an extension of the origin DDPM: q(x (1:…

**Figure 8.** Figure 8: The visualization results of the novel view synthesis. The results show that our method can generate reasonable novel [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 26 canonical work pages

[1]

, " * write output.state after.block = add.period write newline

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...
[2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...
[3]

Barnes, C.; Shechtman, E.; Finkelstein, A.; and Goldman, D. B. 2009. PatchMatch: A randomized correspondence algorithm for structural image editing. ACM Trans. Graph., 28(3): 24

2009
[4]

R.; Nagano, K.; Chan, M

Chan, E. R.; Nagano, K.; Chan, M. A.; Bergman, A. W.; Park, J. J.; Levy, A.; Aittala, M.; De Mello, S.; Karras, T.; and Wetzstein, G. 2023. Generative novel view synthesis with 3d-aware diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 4217--4229

2023
[5]

Chen, A.; Xu, H.; Esposito, S.; Tang, S.; and Geiger, A. 2024 a . LaRa: Efficient Large-Baseline Radiance Fields. arXiv preprint arXiv:2407.04699

arXiv 2024
[6]

Chen, Q.; and Koltun, V. 2017. Photographic image synthesis with cascaded refinement networks. In Proceedings of the IEEE international conference on computer vision, 1511--1520

2017
[7]

Chen, Z.; Wang, Y.; Wang, F.; Wang, Z.; and Liu, H. 2024 b . V3d: Video diffusion models are effective 3d generators. arXiv preprint arXiv:2403.06738

arXiv 2024
[8]

Cheng, W.; Cao, Y.-P.; and Shan, Y. 2023. ID-Pose: Sparse-view Camera Pose Estimation by Inverting Diffusion Models. arXiv preprint arXiv:2306.17140

arXiv 2023

Show all 55 references

[9]

B.; and Vanhoucke, V

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

2022
[10]

T.; and Poole, B

Gao, R.; Holynski, A.; Henzler, P.; Brussee, A.; Martin-Brualla, R.; Srinivasan, P.; Barron, J. T.; and Poole, B. 2024. Cat3d: Create anything in 3d with multi-view diffusion models. arXiv preprint arXiv:2405.10314

2024 arXiv
[11]

Han, X.; Gao, Z.; Kanazawa, A.; Goel, S.; and Gandelsman, Y. 2024. The More You See in 2D the More You Perceive in 3D. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 20912--20922

2024
[12]

Hong, Y.; Zhang, K.; Gu, J.; Bi, S.; Zhou, Y.; Liu, D.; Liu, F.; Sunkavalli, K.; Bui, T.; and Tan, H. 2023. Lrm: Large reconstruction model for single image to 3d. In International Conference on Learning Representations

2023
[13]

Jain, A.; Tancik, M.; and Abbeel, P. 2021. Putting nerf on a diet: Semantically consistent few-shot view synthesis. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 5885--5894

2021
[14]

Jiang, H.; Jiang, Z.; Grauman, K.; and Zhu, Y. 2024 a . Few-view object reconstruction with unknown categories and camera poses. 3DV

2024
[15]

Jiang, H.; Jiang, Z.; Grauman, K.; and Zhu, Y. 2024 b . Few-view object reconstruction with unknown categories and camera poses. In 2024 International Conference on 3D Vision (3DV), 31--41. IEEE

2024
[16]

Jiang, H.; Jiang, Z.; Zhao, Y.; and Huang, Q. 2023. Leap: Liberate sparse-view 3d modeling from camera poses. arXiv preprint arXiv:2310.01410

2023 arXiv
[17]

Kani, B. R. N.; Lee, H.-Y.; Tulyakov, S.; and Tulsiani, S. 2023. UpFusion: Novel View Diffusion from Unposed Sparse View Observations. arXiv preprint arXiv:2312.06661

2023 arXiv
[18]

A.; Ghanem, B.; Tulyakov, S.; and Gilitschenski, I

Kant, Y.; Siarohin, A.; Wu, Z.; Vasilkovsky, M.; Qian, G.; Ren, J.; Guler, R. A.; Ghanem, B.; Tulyakov, S.; and Gilitschenski, I. 2024. SPAD: Spatially Aware Multi-View Diffusers. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 10026--10038

2024
[19]

Kong, X.; Liu, S.; Lyu, X.; Taher, M.; Qi, X.; and Davison, A. J. 2024. EscherNet: A Generative Model for Scalable View Synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 9503--9513

2024
[20]

Laine, S.; Hellsten, J.; Karras, T.; Seol, Y.; Lehtinen, J.; and Aila, T. 2020. Modular Primitives for High-Performance Differentiable Rendering. ACM Transactions on Graphics, 39(6)

2020
[21]

Li, J.; Tan, H.; Zhang, K.; Xu, Z.; Luan, F.; Xu, Y.; Hong, Y.; Sunkavalli, K.; Shakhnarovich, G.; and Bi, S. 2023. Instant3d: Fast text-to-3d with sparse-view generation and large reconstruction model. In International Conference on Learning Representations

2023
[22]

Y.; Ramanan, D.; and Tulsiani, S

Lin, A.; Zhang, J. Y.; Ramanan, D.; and Tulsiani, S. 2024. RelPose++: Recovering 6D Poses from Sparse-view Observations. In 3DV

2024
[23]

Liu, R.; Wu, R.; Van Hoorick, B.; Tokmakov, P.; Zakharov, S.; and Vondrick, C. 2023 a . Zero-1-to-3: Zero-shot One Image to 3D Object. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 9298--9309

2023
[24]

Liu, Y.; Lin, C.; Zeng, Z.; Long, X.; Liu, L.; Komura, T.; and Wang, W. 2023 b . SyncDreamer: Learning to Generate Multiview-consistent Images from a Single-view Image. arXiv preprint arXiv:2309.03453

2023 arXiv
[25]

Long, X.; Guo, Y.-C.; Lin, C.; Liu, Y.; Dou, Z.; Liu, L.; Ma, Y.; Zhang, S.-H.; Habermann, M.; Theobalt, C.; et al. 2024. Wonder3d: Single image to 3d using cross-domain diffusion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 9970--9980

2024
[26]

Long, X.; Lin, C.; Wang, P.; Komura, T.; and Wang, W. 2022. Sparseneus: Fast generalizable neural surface reconstruction from sparse views. In European Conference on Computer Vision, 210--227. Springer

2022
[27]

Loshchilov, I.; and Hutter, F. 2017. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101

2017 arXiv
[28]

L.; and Frahm, J.-M

Schonberger, J. L.; and Frahm, J.-M. 2016. Structure-from-motion revisited. In Proceedings of the IEEE conference on computer vision and pattern recognition, 4104--4113

2016
[29]

Shi, Y.; Wang, P.; Ye, J.; Mai, L.; Li, K.; and Yang, X. 2023. MVDream: Multi-view Diffusion for 3D Generation. arXiv:2308.16512

2023 arXiv
[30]

Y.; Tagliasacchi, A.; Gilitschenski, I.; and Lindell, D

Sinha, S.; Zhang, J. Y.; Tagliasacchi, A.; Gilitschenski, I.; and Lindell, D. B. 2023. Sparsepose: Sparse-view camera pose regression and refinement. In CVPR

2023
[31]

Tang, J.; Chen, Z.; Chen, X.; Wang, T.; Zeng, G.; and Liu, Z. 2024 a . Lgm: Large multi-view gaussian model for high-resolution 3d content creation. arXiv preprint arXiv:2402.05054

2024 arXiv
[32]

Tang, S.; Chen, J.; Wang, D.; Tang, C.; Zhang, F.; Fan, Y.; Chandra, V.; Furukawa, Y.; and Ranjan, R. 2024 b . Mvdiffusion++: A dense high-resolution multi-view diffusion model for single or sparse-view 3d object reconstruction. arXiv preprint arXiv:2402.12712

2024 arXiv
[33]

Tang, S.; Zhang, F.; Chen, J.; Wang, P.; and Yasutaka, F. 2023. MVDiffusion: Enabling Holistic Multi-view Image Generation with Correspondence-Aware Diffusion. arXiv preprint 2307.01097

2023 arXiv
[34]

Tochilkin, D.; Pankratz, D.; Liu, Z.; Huang, Z.; Letts, A.; Li, Y.; Liang, D.; Laforte, C.; Jampani, V.; and Cao, Y.-P. 2024. Triposr: Fast 3d object reconstruction from a single image. arXiv preprint arXiv:2403.02151

2024 arXiv
[35]

Voleti, V.; Yao, C.-H.; Boss, M.; Letts, A.; Pankratz, D.; Tochilkin, D.; Laforte, C.; Rombach, R.; and Jampani, V. 2024. Sv3d: Novel multi-view synthesis and 3d generation from a single image using latent video diffusion. arXiv preprint arXiv:2403.12008

2024 arXiv
[36]

C.; and Liu, Z

Wang, G.; Chen, Z.; Loy, C. C.; and Liu, Z. 2023 a . Sparsenerf: Distilling depth ranking for few-shot novel view synthesis. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 9065--9076

2023
[37]

Wang, J.; Rupprecht, C.; and Novotny, D. 2023. Posediffusion: Solving pose estimation via diffusion-aided bundle adjustment. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 9773--9783

2023
[38]

Wang, P.; Liu, L.; Liu, Y.; Theobalt, C.; Komura, T.; and Wang, W. 2021. NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction. Advances in Neural Information Processing Systems, 34: 27171--27183

2021
[39]

Wang, P.; Tan, H.; Bi, S.; Xu, Y.; Luan, F.; Sunkavalli, K.; Wang, W.; Xu, Z.; and Zhang, K. 2023 b . Pf-lrm: Pose-free large reconstruction model for joint pose and shape prediction. In International Conference on Learning Representations

2023
[40]

Wang, S.; Leroy, V.; Cabon, Y.; Chidlovskii, B.; and Revaud, J. 2024 a . Dust3r: Geometric 3d vision made easy. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 20697--20709

2024
[41]

Wang, Z.; Wang, Y.; Chen, Y.; Xiang, C.; Chen, S.; Yu, D.; Li, C.; Su, H.; and Zhu, J. 2024 b . Crm: Single image to 3d textured mesh with convolutional reconstruction model. arXiv preprint arXiv:2403.05034

2024 arXiv
[43]

Wei, X.; Xiang, F.; Bi, S.; Chen, A.; Sunkavalli, K.; Xu, Z.; and Su, H. 2023 b . Neumanifold: Neural watertight manifold reconstruction with efficient and high-quality rendering support. arXiv preprint arXiv:2305.17134

2023 arXiv
[44]

Wei, X.; Zhang, K.; Bi, S.; Tan, H.; Luan, F.; Deschaintre, V.; Sunkavalli, K.; Su, H.; and Xu, Z. 2024. Meshlrm: Large reconstruction model for high-quality mesh. arXiv preprint arXiv:2404.12385

2024 arXiv
[45]

S.; and Xie, S

Woo, S.; Debnath, S.; Hu, R.; Chen, X.; Liu, Z.; Kweon, I. S.; and Xie, S. 2023. Convnext v2: Co-designing and scaling convnets with masked autoencoders. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 16133--16142

2023
[46]

Wu, C.-H.; Chen, Y.-C.; Solarte, B.; Yuan, L.; and Sun, M. 2023 a . iFusion: Inverting Diffusion for Pose-Free Reconstruction from Sparse Views. arXiv preprint arXiv:2312.17250

2023 arXiv
[47]

P.; Verbin, D.; Barron, J

Wu, R.; Mildenhall, B.; Henzler, P.; Park, K.; Gao, R.; Watson, D.; Srinivasan, P. P.; Verbin, D.; Barron, J. T.; Poole, B.; et al. 2024. Reconfusion: 3d reconstruction with diffusion priors. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition,...

2024
[48]

Wu, T.; Zhang, J.; Fu, X.; Wang, Y.; Ren, J.; Pan, L.; Wu, W.; Yang, L.; Wang, J.; Qian, C.; et al. 2023 b . Omniobject3d: Large-vocabulary 3d object dataset for realistic perception, reconstruction and generation. In Proceedings of the IEEE/CVF Conference on Computer Vision a...

2023
[49]

Yang, J.; Pavone, M.; and Wang, Y. 2023. Freenerf: Improving few-shot neural rendering with free frequency regularization. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 8254--8263

2023
[50]

Yu, A.; Ye, V.; Tancik, M.; and Kanazawa, A. 2021. pixelnerf: Neural radiance fields from one or few images. In CVPR

2021
[51]

Y.; Lin, A.; Kumar, M.; Yang, T.-H.; Ramanan, D.; and Tulsiani, S

Zhang, J. Y.; Lin, A.; Kumar, M.; Yang, T.-H.; Ramanan, D.; and Tulsiani, S. 2024. Cameras as rays: Pose estimation via ray diffusion. arXiv preprint arXiv:2402.14817

2024 arXiv
[52]

Y.; Ramanan, D.; and Tulsiani, S

Zhang, J. Y.; Ramanan, D.; and Tulsiani, S. 2022. Relpose: Predicting probabilistic relative rotation for single objects in the wild. In ECCV

2022
[53]

Zhang, S.; and Zhao, C. 2023. Cross-Video Contextual Knowledge Exploration and Exploitation for Ambiguity Reduction in Weakly Supervised Temporal Action Localization. IEEE Transactions on Circuits and Systems for Video Technology

2023
[54]

Zheng, C.; and Vedaldi, A. 2024. Free3D: Consistent Novel View Synthesis without 3D Representation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 9720--9731

2024
[55]

Zhou, Z.; and Tulsiani, S. 2023. Sparsefusion: Distilling view-conditioned diffusion for 3d reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12588--12597

2023
[56]

Zou, Z.; Cheng, W.; Cao, Y.-P.; Huang, S.-S.; Shan, Y.; and Zhang, S.-H. 2024. Sparse3d: Distilling multiview-consistent diffusion for object reconstruction from sparse views. In Proceedings of the AAAI Conference on Artificial Intelligence, 7900--7908

2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

, " * write output.state after.block = add.period write newline

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

[2] [2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

[3] [3]

Barnes, C.; Shechtman, E.; Finkelstein, A.; and Goldman, D. B. 2009. PatchMatch: A randomized correspondence algorithm for structural image editing. ACM Trans. Graph., 28(3): 24

2009

[4] [4]

R.; Nagano, K.; Chan, M

Chan, E. R.; Nagano, K.; Chan, M. A.; Bergman, A. W.; Park, J. J.; Levy, A.; Aittala, M.; De Mello, S.; Karras, T.; and Wetzstein, G. 2023. Generative novel view synthesis with 3d-aware diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 4217--4229

2023

[5] [5]

Chen, A.; Xu, H.; Esposito, S.; Tang, S.; and Geiger, A. 2024 a . LaRa: Efficient Large-Baseline Radiance Fields. arXiv preprint arXiv:2407.04699

arXiv 2024

[6] [6]

Chen, Q.; and Koltun, V. 2017. Photographic image synthesis with cascaded refinement networks. In Proceedings of the IEEE international conference on computer vision, 1511--1520

2017

[7] [7]

Chen, Z.; Wang, Y.; Wang, F.; Wang, Z.; and Liu, H. 2024 b . V3d: Video diffusion models are effective 3d generators. arXiv preprint arXiv:2403.06738

arXiv 2024

[8] [8]

Cheng, W.; Cao, Y.-P.; and Shan, Y. 2023. ID-Pose: Sparse-view Camera Pose Estimation by Inverting Diffusion Models. arXiv preprint arXiv:2306.17140

arXiv 2023