Pith. sign in

REVIEW 4 major objections 5 minor 48 references

BRUM: Robust 3D Vehicle Reconstruction from 360 Sparse Images

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read From as few as four 360° images of a vehicle, BRUM reconstructs the full 3D shape by synthesizing views along interpolated camera paths and training Gaussian splatting only on high-confidence pixels.

desk verdict Useful dataset and a clean augmentation trick; the 'state-of-the-art' claim is not yet supported by the comparisons. read the letter →

arxiv 2507.12095 v1 pith:7GLCXYME submitted 2025-07-16 cs.CV

Davide Di Nucci , Matteo Tomei , Guido Borghi , Luca Ciuffreda , Roberto Vezzani , Rita Cucchiara This is my paper · ORCID
classification cs.CV
keywords 3Dvehiclereconstructionsparse-viewnovelviewsynthesisGaussiansplattingSLERPaugmentationDUSt3Rposeestimationconfidence-weightedphotometriclossBRUM-dataset360-degreecapture
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 tries to show that accurate 3D vehicle models can be built from as few as four to eight photos arranged around a vehicle, a setting where standard radiance-field and Gaussian-splatting pipelines fail or need dense captures. The proposed method, BRUM, creates extra training views by interpolating camera poses with SLERP and projecting each input image's depth map into those synthetic poses, then trains a Gaussian-splatting model with a loss that ignores pixels that were not reliably re-projected. For real-world captures it replaces the usual structure-from-motion pose estimator with DUSt3R, whose depth and pose estimates are filtered by a confidence threshold. If correct, the method makes vehicle inspection, predictive maintenance, and fleet monitoring practical with quick sparse photo captures instead of dense video. A new public dataset of six synthetic and six real buses is introduced to support this evaluation.

What carries the argument

The load-bearing mechanism is a geometric view-augmentation loop wrapped around Gaussian splatting. For each input camera, SLERP interpolates its pose with its two nearest cameras over small rotation ranges to sample M synthetic poses; the image and depth map are back-projected into a point cloud and rendered into those poses by splatting 3D points onto disks (Eq. 4). An exclusive-NOR mask (Eq. 6) keeps background pixels while removing pixels that did not reproject, and a per-pixel weight (Eqs. 7-8) derived from the splat influence rewards pixels receiving many consistent 3D points. Synthetic views train the Gaussians with a masked, weighted L1 loss only, dropping SSIM because the generated images lack reliable global structure. For real scenes, DUSt3R supplies poses, depth, and per-point confidence, and Eq. 10 prunes the point cloud by a confidence threshold c.

What would settle it

Take a real bus scene, run BRUM with DUSt3R for poses and depth, then rerun it with poses from a surveyed rig and depth from LiDAR or structure-from-motion on dense video while keeping every other setting the same; if the reconstruction quality stays close to the DUSt3R version, the gains are robust to geometry errors, and if it jumps, the reported real-world advantage rests on DUSt3R's accuracy rather than on the augmentation and loss design.

Watch

Extended reading notes

Core claim

The central claim is that sparse-view 360-degree vehicle reconstruction can be made practical by view augmentation plus selective supervision, rather than by a more complex network. With four input views in synthetic scenes and eight in real scenes, BRUM reports the best numbers on the new BRUM-dataset, on CarPatch, and on KRONC across PSNR, SSIM, LPIPS, and AVGE; for example it reaches 25.57 PSNR and 0.040 AVGE on CarPatch under four views, against 23.39 PSNR and 0.056 AVGE for the strongest compared baseline. The improvement comes from giving Gaussian splatting many synthetic views that lie between the real captures, while shielding the optimization from the unreliable pixels those synthetic views contain.

Load-bearing premise

The whole pipeline assumes the depth map and camera pose for each input view are accurate enough that back-projecting and re-rendering produces trustworthy synthetic views; in real scenes that accuracy comes from DUSt3R, so if its depth or pose errors are large the synthetic views feed wrong geometry into the Gaussian optimization.

Editorial extensions

If this is right

  • With only 4-8 images per vehicle, a capture session becomes quick enough for daily inspection rounds and fixed multi-camera setups.
  • Generated views act as free training data, so the method avoids the architectural overhead and extra inference cost of learned sparse-view priors.
  • The selective weighted loss prevents synthetic-view artifacts from corrupting the Gaussian optimization, which is what lets the augmentation help instead of hurt.
  • On denser 12-view inputs the gap to baselines narrows, indicating the method is most valuable precisely in the sparse regime it targets.
  • The released BRUM-dataset gives the vehicle-reconstruction community synthetic and real bus scenes with masks for evaluating sparse-view methods.

Reading between the lines

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

  • Inference: if DUSt3R's depth or pose errors are the limiting factor, coupling the pipeline with pose and depth refinement during Gaussian optimization could push real-world scores closer to the synthetic ones.
  • Inference: the same SLERP-plus-masked-loss recipe may transfer to other object-centric sparse-view tasks, such as furniture, industrial parts, or people, wherever depth and poses can be obtained.
  • Inference: the confidence threshold c is dataset-specific in the paper, so an automatic per-scene selection of c, or a soft weighting based on DUSt3R confidence, is a natural next step.
  • Inference: because the synthetic benchmark uses ground-truth depth and poses, the real-world results should be read as an upper bound on what the augmentation can add when geometry estimation is perfect.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes BRUM, a method for sparse-view 3D vehicle reconstruction that augments Gaussian splatting training with synthetic views. For each input camera, BRUM generates novel poses by SLERP-interpolating toward the two nearest cameras, back-projects the input depth into a point cloud, renders synthetic views via point splatting, and trains 3DGS with a masked, confidence-weighted L1 loss on the generated views while retaining the standard loss on real views. For real-world scenes, DUSt3R replaces COLMAP for camera pose and depth estimation, with a confidence threshold and segmentation masks. The authors also introduce the BRUM-dataset with six synthetic and six real bus scenes, and report improvements over 3DGS, DNGaussians, and SplatFields on the BRUM-dataset, CarPatch, and KRONC.

Significance. If fully substantiated, the contribution is practically valuable: the method is conceptually simple, introduces a new dataset for an under-served domain, and ships code and data publicly. The ablation in Table V gives useful evidence that the masked XNOR operation, the confidence weighting, and the exclusion of SSIM on generated views each contribute. However, the central claim of state-of-the-art performance is currently supported only against three baselines, one of which (DNGaussians) is configured with defaults from an unrelated scene, and the reported numbers are obtained after tuning key hyperparameters on the evaluation datasets. The evidence is therefore not yet strong enough to support the abstract's claim as stated.

major comments (4)
  1. [Sec. II and Tables I-II] The state-of-the-art claim is not established because the authors cite CoherentGS [36], FewViewGS [37], and SC-GS [38] as sparse-view Gaussian splatting methods in Related Work but never compare against them in Tables I or II. These methods are specifically designed for few-view inputs and typically show large gains over plain 3DGS; a missing baseline could exceed BRUM. The authors should either add these or other recent sparse-view GS baselines to the comparison, or explicitly qualify the claim from 'state-of-the-art' to 'state-of-the-art among the compared methods'.
  2. [Sec. IV-C, Tables III-IV] Hyperparameters h and c are selected on the same data used for the final results. Table III tunes the SLERP range h on the Ford scene of KRONC and Table IV tunes the DUSt3R confidence threshold c on bus 1 of the BRUM-dataset; the chosen values are then applied to report the full-dataset numbers in Tables I and II. Because there is no validation split or cross-validation, the reported improvements may reflect test-set tuning. The authors should report results with fixed defaults or perform scene-wise cross-validation, and clarify whether the chosen values generalize to held-out scenes.
  3. [Tables I-II and Fig. 3] The quantitative results are reported as dataset averages without standard deviations or per-scene breakdowns, and Figure 3 plots AVGE curves without error bars. With only six real and six synthetic scenes per dataset, the observed gaps (e.g., +0.85 PSNR over SplatFields on the synthetic BRUM-dataset in Table I) may be within scene-to-scene variability. The authors should report standard deviations, per-scene results, or significance tests so the reader can assess whether the differences are meaningful.
  4. [Sec. III-C and Sec. IV-B vs IV-C] The real-world robustness claim depends on DUSt3R's depth and pose estimates, but the paper provides no direct evaluation of those estimates. The synthetic experiments use ground-truth depth and poses for augmentation (Sec. IV-B), while the real experiments use DUSt3R with a confidence threshold (Sec. III-C) and a reduced SLERP range; the effect of estimation error is not isolated. An ablation that replaces DUSt3R with COLMAP or uses oracle depth/pose on the real scenes would clarify whether BRUM is robust to the preprocessing errors it is designed to handle.
minor comments (5)
  1. [Sec. III-C] There is a typo: 'adpot' should be 'adopt'.
  2. [Sec. IV-C] There is a typo: 'experimentis' should be 'experiments'.
  3. [Eq. (6) and Table V] The symbol ⊕ is defined as XOR, but the operation 1−(V̂ ⊕ V̂) is XNOR; the text calls it 'exclusive nor' and Table V calls it 'XNOR'. Please use a single, explicit notation (e.g., ⊙ for XNOR or write the formula directly) to avoid confusion.
  4. [Sec. IV-C, Table III] The description of h is ambiguous: the text says 'h is fixed at 0.1' while Table III lists h values of 0.05, 0.1, 0.3, 0.5. Clarify whether h denotes the maximum interpolation value or each sampled value, and how the step size 0.025 interacts with these numbers.
  5. [Sec. IV-B and IV-C] Synthetic images are rendered at 800×800 while real images are downsampled to 512×256 for DUSt3R; please state whether the final evaluation was also performed at 512×256 or at the original resolution, as this affects the comparability of metrics across settings.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction in the derivation; the central claim is grounded in held-out test-view evaluation, with only mild in-group benchmark provenance.

full rationale

The paper's derivation chain is self-contained: it takes sparse images plus depth maps and camera poses (Eq. 2), generates novel training views via SLERP interpolation (Eq. 1) and point-cloud splatting (Eqs. 3-4), and trains 3D Gaussians with a weighted masked loss (Eq. 9). The generated views are used only as training augmentation; performance is measured on held-out test views using standard metrics (PSNR, SSIM, LPIPS, AVGE) in Tables I and II. No fitted parameter is renamed as a prediction, and no predicted quantity is defined in terms of a training input. The real-world branch depends on DUSt3R estimates and on tuning the confidence threshold c and interpolation factor h on the same datasets, which is a robustness and evidence-quality concern rather than circularity. The benchmarks CarPatch [3] and KRONC [4] come from the authors' prior work, making the evaluation mildly in-group, but the comparison protocol uses held-out views and standard metrics, so this does not make the central claim circular. Overall, no significant circularity; score 2 reflects the minor self-benchmark provenance, not a circular derivation.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The method relies on the accuracy of depth, pose, and segmentation inputs; in synthetic experiments these are ground truth, which makes the synthetic evaluation favorable. The main tuned quantities are the SLERP range, the DUSt3R confidence threshold, and rendering parameters, all of which are selected empirically and in some cases on the evaluation data itself.

free parameters (5)
  • SLERP interpolation h range and step = step 0.025; h max 0.1 for KRONC, 0.08 for BRUM-dataset, synthetic 0.975
    Controls how far synthetic camera poses move from real ones; chosen empirically, and Table III shows the effect on a single Ford scene.
  • DUSt3R confidence threshold c = 0 for KRONC, 1.5 for BRUM-dataset
    Selected per dataset using Table IV on bus 1; directly affects which points survive to generate views.
  • Point splat radius r = 0.003 synthetic, 0.1 real-world
    Controls projected point size in Eq. 4; hand-set for smoothness.
  • K nearest points per pixel = 16
    Number of points accumulated in z-buffer for rendering; hand-set.
  • Number of input views N = 4 synthetic, 8 real
    They relax the sparse-view constraint on real data because of pose/depth noise; chosen by experiment.
assumptions (6)
  • domain assumption Depth maps for all input views are accurate enough for back-projection (GT in synthetic, DUSt3R in real).
    Used in Sec. III-A Eq. 2; if depth is noisy, generated views are wrong.
  • domain assumption DUSt3R produces usable camera poses and scale-aligned depth for sparse real-world vehicles.
    Replaces COLMAP in Sec. III-C; no independent validation of pose accuracy.
  • domain assumption SLERP interpolation between real camera poses produces plausible object-centric views.
    Used in Eq. 1; assumes static object and smooth camera motion.
  • domain assumption SynSin-style point splatting and z-buffer compositing approximate novel view rendering.
    Adopted from [40] in Sec. III-A; no own evaluation of splatting fidelity.
  • domain assumption Segment Anything masks reliably isolate vehicles in real scenes.
    Used in Sec. III-C Eq. 10 to remove background.
  • domain assumption Standard Gaussian splatting optimization with random point cloud initialization converges from augmented views.
    The synthetic experiments initialize 3DGS with random values and rely on the augmented-view training objective.

how reviews work

0 comments
Cite this review

Pith. "Pith review of BRUM: Robust 3D Vehicle Reconstruction from 360 Sparse Images." pith.science (2026). https://pith.science/paper/7GLCXYME

@misc{pith2026250712095,
  author       = {Pith},
  title        = {Pith review of: BRUM: Robust 3D Vehicle Reconstruction from 360 Sparse Images},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7GLCXYME}},
  note         = {Machine review of arXiv:2507.12095}
}
read the original abstract

Accurate 3D reconstruction of vehicles is vital for applications such as vehicle inspection, predictive maintenance, and urban planning. Existing methods like Neural Radiance Fields and Gaussian Splatting have shown impressive results but remain limited by their reliance on dense input views, which hinders real-world applicability. This paper addresses the challenge of reconstructing vehicles from sparse-view inputs, leveraging depth maps and a robust pose estimation architecture to synthesize novel views and augment training data. Specifically, we enhance Gaussian Splatting by integrating a selective photometric loss, applied only to high-confidence pixels, and replacing standard Structure-from-Motion pipelines with the DUSt3R architecture to improve camera pose estimation. Furthermore, we present a novel dataset featuring both synthetic and real-world public transportation vehicles, enabling extensive evaluation of our approach. Experimental results demonstrate state-of-the-art performance across multiple benchmarks, showcasing the method's ability to achieve high-quality reconstructions even under constrained input conditions.

Figures

Figures reproduced from arXiv: 2507.12095 by the authors.

Figure 1
Figure 1. (a) illustrates an object-centric scene, where ground-truth camera poses are augmented along a SLERP (Sec. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 3
Figure 3. AVGE results by varying number of training images. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Qualitative comparison of 3D reconstruction methods for DNGaussian, 3DGS, SplatFields, and BRUM. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 45 canonical work pages

  1. [36]

    Coherentgs: Sparse novel view synthesis with coherent 3d gaussians,

    A. Paliwal, W. Ye, J. Xiong, D. Kotovenko, R. Ranjan, V . Chandra, and N. K. Kalantari, “Coherentgs: Sparse novel view synthesis with coherent 3d gaussians,” in Eur . Conf. Comput. Vis., 2024

    work page 2024

  2. [37]

    Fewviewgs: Gaussian splatting with few view matching and multi-stage training,

    R. Yin, V . Yugay, Y . Li, S. Karaoglu, and T. Gevers, “Fewviewgs: Gaussian splatting with few view matching and multi-stage training,” in Adv. Neural Inform. Process. Syst. , 2024

    work page 2024

  3. [38]

    Structure consistent gaussian splatting with matching prior for few-shot novel view synthesis,

    R. Peng, W. Xu, L. Tang, L. Liao, J. Jiao, and R. Wang, “Structure consistent gaussian splatting with matching prior for few-shot novel view synthesis,” in Adv. Neural Inform. Process. Syst. , 2024

    work page 2024

  4. [1]

    Nerf: Representing scenes as neural radiance fields for view synthesis,

    B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoor- thi, and R. Ng, “Nerf: Representing scenes as neural radiance fields for view synthesis,” Communications of the ACM , 2021

    work page 2021

  5. [2]

    3d gaussian splatting for real-time radiance field rendering

    B. Kerbl, G. Kopanas, T. Leimk ¨uhler, and G. Drettakis, “3d gaussian splatting for real-time radiance field rendering.” ACM Trans. Graph. , 2023

    work page 2023

  6. [3]

    Carpatch: A synthetic benchmark for radiance field evaluation on vehicle components,

    D. Di Nucci, A. Simoni, M. Tomei, L. Ciuffreda, R. Vezzani, and R. Cucchiara, “Carpatch: A synthetic benchmark for radiance field evaluation on vehicle components,” in International Conference on Image Analysis and Processing , 2023

    work page 2023

  7. [4]

    Kronc: Keypoint-based robust camera optimization for 3d car reconstruction,

    ——, “Kronc: Keypoint-based robust camera optimization for 3d car reconstruction,” in Eur . Conf. Comput. Vis. Worksh., 2024

    work page 2024

  8. [5]

    Neurad: Neural rendering for autonomous driving,

    A. Tonderski, C. Lindstr ¨om, G. Hess, W. Ljungbergh, L. Svensson, and C. Petersson, “Neurad: Neural rendering for autonomous driving,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2024

    work page 2024

Show all 48 references
  1. [6]

    Driv- inggaussian: Composite gaussian splatting for surrounding dynamic autonomous driving scenes,

    X. Zhou, Z. Lin, X. Shan, Y . Wang, D. Sun, and M.-H. Yang, “Driv- inggaussian: Composite gaussian splatting for surrounding dynamic autonomous driving scenes,” in IEEE Conf. Comput. Vis. Pattern Recog., 2024

    2024

  2. [7]

    Mega-nerf: Scalable construction of large-scale nerfs for virtual fly-throughs,

    H. Turki, D. Ramanan, and M. Satyanarayanan, “Mega-nerf: Scalable construction of large-scale nerfs for virtual fly-throughs,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2022

    2022

  3. [8]

    Citydreamer: Compositional generative model of unbounded 3d cities,

    H. Xie, Z. Chen, F. Hong, and Z. Liu, “Citydreamer: Compositional generative model of unbounded 3d cities,” in IEEE Conf. Comput. Vis. Pattern Recog., 2024

    2024

  4. [9]

    Citygaussian: Real-time high-quality large-scale scene rendering with gaussians,

    Y . Liu, C. Luo, L. Fan, N. Wang, J. Peng, and Z. Zhang, “Citygaussian: Real-time high-quality large-scale scene rendering with gaussians,” in Eur . Conf. Comput. Vis., 2024

    2024

  5. [10]

    Structure-from-motion revisited,

    J. L. Schonberger and J.-M. Frahm, “Structure-from-motion revisited,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2016

    2016

  6. [11]

    Sparf: Neural radiance fields from sparse and noisy poses,

    P. Truong, M.-J. Rakotosaona, F. Manhardt, and F. Tombari, “Sparf: Neural radiance fields from sparse and noisy poses,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2023

    2023

  7. [12]

    Dust3r: Geometric 3d vision made easy,

    S. Wang, V . Leroy, Y . Cabon, B. Chidlovskii, and J. Revaud, “Dust3r: Geometric 3d vision made easy,” in IEEE Conf. Comput. Vis. Pattern Recog., 2024

    2024

  8. [13]

    Instant neural graphics primitives with a multiresolution hash encoding,

    T. M ¨uller, A. Evans, C. Schied, and A. Keller, “Instant neural graphics primitives with a multiresolution hash encoding,” ACM Trans. Graph., 2022

    2022

  9. [14]

    Plenoxels: Radiance fields without neural networks,

    S. Fridovich-Keil, A. Yu, M. Tancik, Q. Chen, B. Recht, and A. Kanazawa, “Plenoxels: Radiance fields without neural networks,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2022

    2022

  10. [15]

    Barf: Bundle- adjusting neural radiance fields,

    C.-H. Lin, W.-C. Ma, A. Torralba, and S. Lucey, “Barf: Bundle- adjusting neural radiance fields,” in Int. Conf. Comput. Vis. , 2021

    2021

  11. [16]

    Local-to-global registration for bundle-adjusting neural radiance fields,

    Y . Chen, X. Chen, X. Wang, Q. Zhang, Y . Guo, Y . Shan, and F. Wang, “Local-to-global registration for bundle-adjusting neural radiance fields,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2023

    2023

  12. [17]

    Nope-nerf: Optimising neural radiance field with no pose prior,

    W. Bian, Z. Wang, K. Li, J.-W. Bian, and V . A. Prisacariu, “Nope-nerf: Optimising neural radiance field with no pose prior,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2023

    2023

  13. [18]

    Neural refinement for absolute pose regression with fea- ture synthesis,

    S. Chen, Y . Bhalgat, X. Li, J.-W. Bian, K. Li, Z. Wang, and V . A. Prisacariu, “Neural refinement for absolute pose regression with fea- ture synthesis,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2024

    2024

  14. [19]

    Compact 3d scene representation via self-organizing gaussian grids,

    W. Morgenstern, F. Barthel, A. Hilsmann, and P. Eisert, “Compact 3d scene representation via self-organizing gaussian grids,” in Eur . Conf. Comput. Vis., 2024

    2024

  15. [20]

    Compressed 3d gaussian splatting for accelerated novel view synthesis,

    S. Niedermayr, J. Stumpfegger, and R. Westermann, “Compressed 3d gaussian splatting for accelerated novel view synthesis,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2024

    2024

  16. [21]

    Lightgaus- sian: Unbounded 3d gaussian compression with 15x reduction and 200+ fps,

    Z. Fan, K. Wang, K. Wen, Z. Zhu, D. Xu, and Z. Wang, “Lightgaus- sian: Unbounded 3d gaussian compression with 15x reduction and 200+ fps,” in Adv. Neural Inform. Process. Syst. , 2024

    2024

  17. [22]

    Gaussianimage: 1000 fps image representation and compression by 2d gaussian splatting,

    X. Zhang, X. Ge, T. Xu, D. He, Y . Wang, H. Qin, G. Lu, J. Geng, and J. Zhang, “Gaussianimage: 1000 fps image representation and compression by 2d gaussian splatting,” in Eur . Conf. Comput. Vis. , 2024

    2024

  18. [23]

    Segment anything in 3d with nerfs,

    J. Cen, Z. Zhou, J. Fang, W. Shen, L. Xie, D. Jiang, X. Zhang, Q. Tian et al. , “Segment anything in 3d with nerfs,” in Adv. Neural Inform. Process. Syst., 2023

    2023

  19. [24]

    Semantic anything in 3d gaussians,

    X. Hu, Y . Wang, L. Fan, J. Fan, J. Peng, Z. Lei, Q. Li, and Z. Zhang, “Semantic anything in 3d gaussians,” arXiv:2401.17857, 2024

    2024 arXiv

  20. [25]

    Gaussianeditor: Swift and controllable 3d editing with gaussian splatting,

    Y . Chen, Z. Chen, C. Zhang, F. Wang, X. Yang, Y . Wang, Z. Cai, L. Yang, H. Liu, and G. Lin, “Gaussianeditor: Swift and controllable 3d editing with gaussian splatting,” in IEEE Conf. Comput. Vis. Pattern Recog., 2024

    2024

  21. [26]

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

    B. Poole, A. Jain, J. T. Barron, and B. Mildenhall, “Dreamfusion: Text-to-3d using 2d diffusion,” arXiv:2209.14988, 2022

    2022 arXiv

  22. [27]

    Dreamscene360: Unconstrained text-to- 3d scene generation with panoramic gaussian splatting,

    S. Zhou, Z. Fan, D. Xu, H. Chang, P. Chari, T. Bharadwaj, S. You, Z. Wang, and A. Kadambi, “Dreamscene360: Unconstrained text-to- 3d scene generation with panoramic gaussian splatting,” in Eur . Conf. Comput. Vis., 2024

    2024

  23. [28]

    Text-to-3d using gaussian splatting,

    Z. Chen, F. Wang, Y . Wang, and H. Liu, “Text-to-3d using gaussian splatting,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2024

    2024

  24. [29]

    Luciddreamer: Towards high-fidelity text-to-3d generation via interval score match- ing,

    Y . Liang, X. Yang, J. Lin, H. Li, X. Xu, and Y . Chen, “Luciddreamer: Towards high-fidelity text-to-3d generation via interval score match- ing,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2024

    2024

  25. [30]

    pixelnerf: Neural radiance fields from one or few images,

    A. Yu, V . Ye, M. Tancik, and A. Kanazawa, “pixelnerf: Neural radiance fields from one or few images,” in IEEE Conf. Comput. Vis. Pattern Recog., 2021

    2021

  26. [31]

    Ibrnet: Learning multi-view image-based rendering,

    Q. Wang, Z. Wang, K. Genova, P. P. Srinivasan, H. Zhou, J. T. Barron, R. Martin-Brualla, N. Snavely, and T. Funkhouser, “Ibrnet: Learning multi-view image-based rendering,” in IEEE Conf. Comput. Vis. Pattern Recog., 2021

    2021

  27. [32]

    Corresnerf: Image corre- spondence priors for neural radiance fields,

    Y . Lao, X. Xu, X. Liu, H. Zhao et al. , “Corresnerf: Image corre- spondence priors for neural radiance fields,” in Adv. Neural Inform. Process. Syst., 2023

    2023

  28. [33]

    Depth-supervised nerf: Fewer views and faster training for free,

    K. Deng, A. Liu, J.-Y . Zhu, and D. Ramanan, “Depth-supervised nerf: Fewer views and faster training for free,” in IEEE Conf. Comput. Vis. Pattern Recog., 2022

    2022

  29. [34]

    Dense depth priors for neural radiance fields from sparse input views,

    B. Roessle, J. T. Barron, B. Mildenhall, P. P. Srinivasan, and M. Nießner, “Dense depth priors for neural radiance fields from sparse input views,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2022

    2022

  30. [35]

    Splatfields: Neural gaussian splats for sparse 3d and 4d reconstruction,

    M. Mihajlovic, S. Prokudin, S. Tang, R. Maier, F. Bogo, T. Tung, and E. Boyer, “Splatfields: Neural gaussian splats for sparse 3d and 4d reconstruction,” in Eur . Conf. Comput. Vis., 2024

    2024

  31. [39]

    Animating rotation with quaternion curves,

    K. Shoemake, “Animating rotation with quaternion curves,” in Pro- ceedings of the 12th annual conference on Computer graphics and interactive techniques, 1985, pp. 245–254

    1985

  32. [40]

    Synsin: End-to- end view synthesis from a single image,

    O. Wiles, G. Gkioxari, R. Szeliski, and J. Johnson, “Synsin: End-to- end view synthesis from a single image,” in IEEE Conf. Comput. Vis. Pattern Recog., 2020

    2020

  33. [41]

    Image quality assessment: from error visibility to structural similarity,

    Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Trans. Image Process. , 2004

    2004

  34. [42]

    An image is worth 16x16 words: Transformers for image recognition at scale,

    A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly et al., “An image is worth 16x16 words: Transformers for image recognition at scale,” arXiv:2010.11929, 2020

    2010 arXiv

  35. [43]

    Attention is all you need,

    A. Vaswani, “Attention is all you need,” Adv. Neural Inform. Process. Syst., 2017

    2017

  36. [44]

    Segment anything,

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y . Lo et al. , “Segment anything,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2023

    2023

  37. [45]

    Dngaussian: Optimizing sparse-view 3d gaussian radiance fields with global-local depth normalization,

    J. Li, J. Zhang, X. Bai, J. Zheng, X. Ning, J. Zhou, and L. Gu, “Dngaussian: Optimizing sparse-view 3d gaussian radiance fields with global-local depth normalization,” in IEEE Conf. Comput. Vis. Pattern Recog., 2024

    2024

  38. [46]

    The unreasonable effectiveness of deep features as a perceptual metric,

    R. Zhang, P. Isola, A. A. Efros, E. Shechtman, and O. Wang, “The unreasonable effectiveness of deep features as a perceptual metric,” in IEEE Conf. Comput. Vis. Pattern Recog. , 2018

    2018

  39. [47]

    Regnerf: Regularizing neural radiance fields for view synthesis from sparse inputs,

    M. Niemeyer, J. T. Barron, B. Mildenhall, M. S. Sajjadi, A. Geiger, and N. Radwan, “Regnerf: Regularizing neural radiance fields for view synthesis from sparse inputs,” in IEEE Conf. Comput. Vis. Pattern Recog., 2022

    2022

  40. [48]

    Vision transformers for dense prediction,

    R. Ranftl, A. Bochkovskiy, and V . Koltun, “Vision transformers for dense prediction,” in Int. Conf. Comput. Vis. , 2021

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.