Pith. sign in

REVIEW 4 major objections 5 minor 38 references

T-Graph: Enhancing Sparse-view Camera Pose Estimation by Pairwise Translation Graph

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read Pairwise translation supervision, organized as a fully connected graph, improves sparse-view camera pose estimation.

desk verdict A lightweight training-only pairwise translation module that seems to help sparse-view pose estimation, but the main RelPose++/CO3D comparison relies on borrowed baselines and no error bars, so treat the headline gain as plausible rather than proven. read the letter →

arxiv 2505.01207 v1 pith:GFPNRVFW submitted 2025-05-02 cs.CV

Qingyu Xian , Weiqin Jiao , Hao Cheng , Berend Jan van der Zwaag , Yanqiu Huang This is my paper · ORCID
classification cs.CV
keywords cameraposeestimationsparse-viewpairwisetranslationgraphrelative-tpair-tauxiliarysupervisionplug-and-playmodule
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 claims that sparse-view camera pose estimation improves when the model is also supervised on the translation between every pair of cameras, not just each camera's translation to a fixed world origin. The proposed T-Graph is a lightweight module that shares the baseline feature extractor, predicts pairwise translations through a small MLP, and arranges them as a fully connected graph with cameras as nodes. It is active only during training and removed at inference, so any gains come without slowing the deployed model. Across the RelPose++ and Forge-2D baselines on the CO3D and IMC PhotoTourism datasets, the authors report consistent gains, with camera center accuracy improving by 1% to 6% from 2 to 8 viewpoints.

What carries the argument

The load-bearing mechanism is the fully connected translation graph: for $n$ cameras there are $\binom{n}{2}$ edges, and each edge is produced by the same lightweight MLP regressor from the pair of image features $(f_i, f_j)$. The graph replaces sparse per-camera translation regression with dense pairwise supervision, and the scaling factors $k_1 = n / (2 \binom{n}{2})$ for pair-t and $k_2 = n / \binom{n}{2}$ for relative-t keep the auxiliary loss comparable to the baseline translation loss. The two representations matter because pair-t removes the rotation-translation coupling by putting the origin at the optical-axis intersection, while relative-t keeps a simple local frame that remains stable when optical axes are nearly parallel. Because the module is dropped at inference, it can only improve accuracy by shaping the shared feature extractor during training.

What would settle it

Retrain RelPose++ from scratch on CO3D with the exact hyperparameters, batch sizes, and random seeds used for the T-Graph runs, then compare camera center accuracy at 3 to 8 views; if the gap between baseline and T-Graph disappears or falls below 1%, the claimed improvement is at least partly an artifact of training conditions.

Watch

Extended reading notes

Core claim

The paper's central claim is that pairwise translation information is a valuable and underused supervision signal for sparse-view pose estimation. T-Graph builds a complete graph whose nodes are cameras and whose edges are translations between camera pairs, predicted by a shared MLP from paired image features. These pairwise predictions are not used as final output; they contribute an L1 loss during training that steers the shared feature extractor toward globally informative representations. Two edge encodings are introduced: relative-t, which sets the world origin at one camera and remains entangled with rotation, and pair-t, which sets the origin at the intersection of the two optical axes and thereby disentangles translation from rotation. Empirically, pair-t suits center-facing camera configurations such as CO3D, while relative-t suits roughly parallel configurations such as IMC PhotoTourism.

Load-bearing premise

The comparison on CO3D assumes the published RelPose++ numbers were obtained under the same training conditions as the authors' runs, since those baselines were not retrained; if the training settings differ, the reported improvements could be inflated.

Editorial extensions

If this is right

  • Camera center accuracy improves by 1% to 6% across 2 to 8 viewpoints on both CO3D and IMC PhotoTourism when T-Graph is added.
  • The gains transfer across two architecturally distinct baselines, RelPose++ (energy-based rotation model) and Forge-2D (purely discriminative), which supports the module's generalizability.
  • The choice of representation should follow the geometry of the camera distribution: pair-t for center-facing cameras, relative-t for mostly parallel cameras.
  • Because T-Graph is removed at inference, it adds no runtime cost and only modest parameter growth (+5% for RelPose++, +11% for Forge-2D).
  • Pairwise ground-truth translations are a cost-effective additional training signal for sparse-view pose estimation.

Reading between the lines

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

  • The same pairwise-supervision idea could be applied to relative rotations, producing a fully connected rotation graph that may complement the translation graph and improve pose consistency further.
  • A learned combination or soft switch between relative-t and pair-t could remove the need to choose the representation manually for each dataset.
  • Since the module is removed at inference, any benefit must be mediated through the shared feature extractor; this makes T-Graph a useful probe for whether explicit geometric supervision during training produces more pose-aware features, a claim that could be checked by probing feature similarity.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes T-Graph, a lightweight, plug-and-play auxiliary branch for sparse-view camera pose estimation. During training, T-Graph takes pairs of image features, predicts pairwise translations through a shared MLP, and adds an L1 loss under either a relative-t or a pair-t representation; the branch is removed at inference, so the baseline architecture is unchanged. The method is integrated into RelPose++ and a 2D-branch variant of Forge and evaluated on CO3D and IMC PhotoTourism for 2 to 8 viewpoints. The authors report consistent improvements across the four experiment groups, with camera-center accuracy improving by 1% to 6%, and conclude that pairwise translation supervision is a cost-effective way to improve pose estimation.

Significance. If the claimed gains hold, T-Graph would be a useful training-time regularizer for sparse-view pose estimation: it adds only a small number of parameters, does not change inference cost, and is architecturally agnostic. The paper also proposes two pairwise translation representations with a plausible geometric interpretation, which is a useful conceptual contribution for choosing supervision under different camera configurations. The empirical evaluation covers two public datasets and two representative baseline families, and the qualitative visualizations support the general direction. However, the strongest quantitative claim is not yet rigorously established because the headline CO3D/RelPose++ comparison relies on published numbers rather than retrained baselines, error bars are absent, and the added MLP capacity is not ablated. The paper is clearly written and the idea is simple and easy to reproduce, but the evidence as presented supports a moderate, not a strong, claim of consistent improvement.

major comments (4)
  1. [Sec. 4.1, Table 2] The comparison between RelPose++ and T-Graph on CO3D is not a controlled experiment. The manuscript states that the RelPose++ evaluation results are taken directly from the original publication and that no additional training was conducted, while the T-Graph variants are trained on a single H100 with batch size 22. Different hardware, batch size, effective epoch count, learning-rate schedule, or random seed in the published baseline could account for part or all of the reported gains. The authors should retrain RelPose++ under their exact pipeline and report those numbers, or otherwise the 1–6% headline improvement in camera-center accuracy should be substantially qualified.
  2. [Sec. 4.1, Tables 2–5] The results are reported without error bars or multiple seeds; for EG 2–4 the paper explicitly uses a fixed random seed and a single random sampling trial per viewpoint count. Many of the reported improvements are small (typically 0.2–2 percentage points), and some entries regress relative to the baseline, for example Table 3 translation accuracy at 2 views (0.589 vs. 0.595) and Table 2 rotation accuracy at 2 views for relative-t (0.684 vs. 0.698). The claim of 'consistent improvements' in Sec. 4.2 and the abstract therefore goes beyond what the reported numbers establish. The authors should provide variance across seeds and sampling trials, or temper the claim accordingly.
  3. [Sec. 3.1, Table 6] There is no ablation that controls for the added capacity of the 6-layer MLP branch. Table 6 shows that T-Graph increases model size by 5–11%, and since the branch is active only during training, the observed gains could partly come from extra model capacity, the additional training signal in general, or regularization effects. An ablation that replaces the pairwise-translation loss with an auxiliary loss of matched capacity (for example, the same MLP predicting scalar targets or per-camera absolute translations) would be needed to attribute the improvement specifically to pairwise translation supervision.
  4. [Sec. 3.2, Fig. 3(b)] The pair-t formulation is geometrically inaccurate for the general case used in the experiments. The paper defines W_k as the point that minimizes the distance to both optical axes, but then asserts that its coordinates in the two camera frames are (0,0,D_i) and (0,0,D_j). When the optical axes do not exactly intersect, which is the generic case and is explicitly contemplated in the 'approximately intersect' discussion in Sec. 5, W_k is not on either optical axis, so the vector from a camera center to W_k has off-axis components and is not simply (0,0,D). The rotation-disentanglement claim should be restated, or the representation should be redefined (e.g., using the foot of the perpendicular from W_k onto each axis) so that the stated coordinates are correct.
minor comments (5)
  1. [Abstract, Sec. 4.1] The dataset name is written as 'C03D' in the abstract and in Sec. 4.1; this should be 'CO3D' throughout.
  2. [Throughout] The paper alternates between 'T-Graph' and 'T-graph' (for example, the abstract uses both). Please use a single consistent name.
  3. [Eq. (1)] The notation in Eq. (1) is not fully explained: X^w is called 'the point in the world frame' and X^c_j 'the corresponding point in the camera frame of C_j', but the relation between these points and the translation vector t_{i→j} should be stated more explicitly, including the coordinate-frame definitions for R_{i→j}.
  4. [Sec. 4.1] The description of the Forge-2D integration is somewhat hard to follow: the text says self-attention is applied to the first image feature so that all k pose features can serve as inputs to T-Graph. A short pseudo-code or tensor-shape description would clarify how the k pose features are obtained and used.
  5. [Sec. 4.2] The statement that 'the camera center accuracy at a threshold of 0.2 is always 1 when the number of views is 2' is given without justification. A one-sentence explanation (for example, that two camera centers can always be aligned by a similarity transform) would help the reader.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: T-Graph's auxiliary pairwise translation loss is supervised by external COLMAP ground-truth poses and introduces no fitted parameter that is later reported as a prediction; the unretrained RelPose++/CO3D baseline is an experimental-control caveat, not a derivation-circularity issue.

full rationale

The central claim is that adding a pairwise-translation supervision branch improves sparse-view pose estimation. The additional loss LT-Graph (Eq. 2) compares T-Graph outputs with ground-truth pairwise translations, either relative-t or pair-t, computed from the dataset's COLMAP-derived poses. These targets are external to the model: they are not derived from the baseline's predictions, and no parameter of T-Graph is fitted to the reported evaluation metrics. The two representations are defined geometrically in Sec. 3.2 (Eq. 1 for relative-t, and the optical-axis intersection construction for pair-t), so the supervision signal is not a re-packaging of the baseline loss or the evaluation criterion. The scaling factors k1 and k2 in Eq. (3) are fixed counts used only to keep loss magnitudes comparable; they are not learned from the test results. The paper is self-contained in its derivation: the method is a straightforward auxiliary supervised branch, removed at inference, and the claimed gains are empirical comparisons against public baselines on public datasets. The only notable weakness is that the RelPose++/CO3D numbers in Table 2 are taken from the original publication rather than retrained under the authors' exact pipeline, as explicitly stated in Sec. 4.1. That is a legitimate experimental-control concern about whether the comparison conditions are identical, but it is not a circularity: the baseline numbers are independent external results, and Tables 3-5 do train the corresponding baselines in-house. No self-citation chain carries the argument, no uniqueness theorem is imported from the authors' prior work, and no fitted value is renamed as a prediction. Therefore the appropriate circularity score is 0.

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

The claims rest on standard dataset assumptions and geometric definitions, not on fitted constants. Loss scaling factors are derived from output counts rather than tuned to data.

assumptions (4)
  • domain assumption Ground-truth camera poses from COLMAP used in CO3D and IMC PhotoTourism are accurate enough to serve as supervision for pairwise translation targets.
    The paper computes pairwise translation ground truths from these poses (Sec 3.3); if the poses contain significant error, the auxiliary targets are noisy.
  • domain assumption The intersection point of two camera optical axes (used in pair-t) is defined as the 3D point minimizing distance to both axes, and it is finite and stable for the datasets where pair-t is applied.
    Sec 3.2 defines W_k implicitly; the paper notes this point becomes unstable for near-parallel cameras (IMC PhotoTourism), so its stability is assumed for CO3D-like configurations.
  • standard math An optimal similarity transform (Umeyama least-squares) between predicted and ground-truth camera centers provides a valid evaluation of camera center accuracy.
    Sec 4.1, following RelPose++.
  • domain assumption The features extracted by the baseline backbones are informative enough for the T-Graph MLP to regress pairwise translations, and the resulting gradients improve the shared feature extractor.
    Sec 3.1 design assumption; the experiments support it but it is not independently proven.

how reviews work

0 comments
Cite this review

Pith. "Pith review of T-Graph: Enhancing Sparse-view Camera Pose Estimation by Pairwise Translation Graph." pith.science (2026). https://pith.science/paper/GFPNRVFW

@misc{pith2026250501207,
  author       = {Pith},
  title        = {Pith review of: T-Graph: Enhancing Sparse-view Camera Pose Estimation by Pairwise Translation Graph},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GFPNRVFW}},
  note         = {Machine review of arXiv:2505.01207}
}
read the original abstract

Sparse-view camera pose estimation, which aims to estimate the 6-Degree-of-Freedom (6-DoF) poses from a limited number of images captured from different viewpoints, is a fundamental yet challenging problem in remote sensing applications. Existing methods often overlook the translation information between each pair of viewpoints, leading to suboptimal performance in sparse-view scenarios. To address this limitation, we introduce T-Graph, a lightweight, plug-and-play module to enhance camera pose estimation in sparse-view settings. T-graph takes paired image features as input and maps them through a Multilayer Perceptron (MLP). It then constructs a fully connected translation graph, where nodes represent cameras and edges encode their translation relationships. It can be seamlessly integrated into existing models as an additional branch in parallel with the original prediction, maintaining efficiency and ease of use. Furthermore, we introduce two pairwise translation representations, relative-t and pair-t, formulated under different local coordinate systems. While relative-t captures intuitive spatial relationships, pair-t offers a rotation-disentangled alternative. The two representations contribute to enhanced adaptability across diverse application scenarios, further improving our module's robustness. Extensive experiments on two state-of-the-art methods (RelPose++ and Forge) using public datasets (C03D and IMC PhotoTourism) validate both the effectiveness and generalizability of T-Graph. The results demonstrate consistent improvements across various metrics, notably camera center accuracy, which improves by 1% to 6% from 2 to 8 viewpoints.

Figures

Figures reproduced from arXiv: 2505.01207 by the authors.

Figure 1
Figure 1. Illustration of two pairwise translation representations based on different coor [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Comparison of camera pose estimation architectures with and without T-Graph. [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. T-Graph (simplified with four cameras) module with two different pairwise [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: RelPose++ [15] combined with T-Graph. RelPose++ adopts a ResNet-50 back [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]
Figure 5
Figure 5. Figure 5: Forge-2D [19] combined with T-Graph. Forge-2D employs a multi-layer CNN [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Visualization of recovered camera poses on CO3D. Ground truth, original pre [PITH_FULL_IMAGE:figures/full_fig_p020_6.png]
Figure 7
Figure 7. Figure 7: Visualization of recovered camera poses on IMC PhotoTourism. Ground truth, [PITH_FULL_IMAGE:figures/full_fig_p021_7.png]
Figure 8
Figure 8. Figure 8: Camera pose visualization of different scenarios. (a) Center-facing cameras, [PITH_FULL_IMAGE:figures/full_fig_p023_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 33 canonical work pages

  1. [1]

    S. Wang, D. Quan, X. Liang, M. Ning, Y. Guo, L. Jiao, A deep learn- ing framework for remote sensing image registration, ISPRS Journal of Photogrammetry and Remote Sensing 145 (2018) 148–164

    work page 2018

  2. [2]

    Q. Yu, D. Ni, Y. Jiang, Y. Yan, J. An, T. Sun, Universal sar and optical image registration via a novel sift framework based on nonlinear diffusion and a polar spatial-frequency descriptor, ISPRS Journal of Photogram- metry and Remote Sensing 171 (2021) 1–17

    work page 2021

  3. [3]

    K. Chen, H. Yu, W. Yang, L. Yu, S. Scherer, G.-S. Xia, I2d-loc: Cam- era localization via image to lidar depth flow, ISPRS Journal of Pho- togrammetry and Remote Sensing 194 (2022) 209–221. doi:https: //doi.org/10.1016/j.isprsjprs.2022.10.009

    work page doi:10.1016/j.isprsjprs.2022.10.009 2022

  4. [4]

    D. Wang, J. Wang, Y. Tian, Y. Fang, Z. Yuan, M. Xu, Pal-slam2: Visual and visual–inertial monocular slam for panoramic annular lens, ISPRS Journal of Photogrammetry and Remote Sensing 211 (2024) 35–

    work page 2024

  5. [5]

    D. Yao, M. Zhu, H. Zhu, W. Cai, L. Zhou, Integrating synthetic datasets with clip semantic insights for single image localization advancements, ISPRS Journal of Photogrammetry and Remote Sensing 218 (2024) 198–

    work page 2024

  6. [6]

    Stucker, K

    C. Stucker, K. Schindler, Resdepth: A deep residual prior for 3d recon- struction from high-resolution satellite images, ISPRS Journal of Pho- togrammetry and Remote Sensing 183 (2022) 560–580

    work page 2022

  7. [7]

    D. Yu, S. Ji, J. Liu, S. Wei, Automatic 3d building reconstruction from multi-view aerial images with deep learning, ISPRS Journal of Pho- togrammetry and Remote Sensing 171 (2021) 155–170

    work page 2021

  8. [8]

    Z. Li, J. Shan, Ransac-based multi primitive building reconstruction from 3d point clouds, ISPRS Journal of Photogrammetry and Remote Sensing 185 (2022) 247–260

    work page 2022

Show all 38 references
  1. [9]

    J. Gao, J. Liu, S. Ji, A general deep learning based framework for 3d reconstruction from multi-view stereo satellite images, ISPRS Journal of Photogrammetry and Remote Sensing 195 (2023) 446–461. 25

    2023

  2. [10]

    Worden, C

    K. Worden, C. R. Farrar, G. Manson, G. Park, The fundamental ax- ioms of structural health monitoring, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences 463 (2082) (2007) 1639–1664

    2007

  3. [11]

    Sturm, N

    J. Sturm, N. Engelhard, F. Endres, W. Burgard, D. Cremers, A bench- mark for the evaluation of rgb-d slam systems, in: 2012 IEEE/RSJ international conference on intelligent robots and systems, IEEE, 2012, pp. 573–580

    2012

  4. [12]

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

    2016

  5. [13]

    W. Kehl, F. Manhardt, F. Tombari, S. Ilic, N. Navab, Ssd-6d: Making rgb-based 3d detection and 6d pose estimation great again, in: Proceed- ings of the IEEE international conference on computer vision, 2017, pp. 1521–1529

    2017

  6. [14]

    Xiang, T

    Y. Xiang, T. Schmidt, V. Narayanan, D. Fox, Posecnn: A convolutional neural network for 6d object pose estimation in cluttered scenes, arXiv preprint arXiv:1711.00199 (2017)

    2017 arXiv

  7. [15]

    A. Lin, J. Y. Zhang, D. Ramanan, S. Tulsiani, Relpose++: Recovering 6d poses from sparse-view observations, in: 2024 International Confer- ence on 3D Vision (3DV), IEEE, 2024, pp. 106–115

    2024

  8. [16]

    Sinha, J

    S. Sinha, J. Y. Zhang, A. Tagliasacchi, I. Gilitschenski, D. B. Lindell, Sparsepose: Sparse-view camera pose regression and refinement, in: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 21349–21359

    2023

  9. [17]

    J. Wang, C. Rupprecht, D. Novotny, Posediffusion: Solving pose esti- mation via diffusion-aided bundle adjustment, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 9773–9783

    2023

  10. [18]

    J. Y. Zhang, A. Lin, M. Kumar, T.-H. Yang, D. Ramanan, S. Tul- siani, Cameras as rays: Pose estimation via ray diffusion, arXiv preprint arXiv:2402.14817 (2024). 26

    2024 arXiv

  11. [19]

    Jiang, Z

    H. Jiang, Z. Jiang, K. Grauman, Y. Zhu, Few-view object reconstruc- tion with unknown categories and camera poses, in: 2024 International Conference on 3D Vision (3DV), IEEE, 2024, pp. 31–41

    2024

  12. [20]

    Reizenstein, R

    J. Reizenstein, R. Shapovalov, P. Henzler, L. Sbordone, P. Labatut, D. Novotny, Common objects in 3d: Large-scale learning and evaluation of real-life 3d category reconstruction, in: Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 10901–10911

    2021

  13. [21]

    Y. Jin, D. Mishkin, A. Mishchuk, J. Matas, P. Fua, K. M. Yi, E. Trulls, Image matching across wide baselines: From paper to practice, Interna- tional Journal of Computer Vision 129 (2) (2021) 517–547

    2021

  14. [22]

    D. G. Lowe, Distinctive image features from scale-invariant keypoints, International journal of computer vision 60 (2004) 91–110

    2004

  15. [23]

    M. A. Fischler, R. C. Bolles, A paradigm for model fitting with ap- plications to image analysis and automated cartography (reprinted in readings in computer vision, ed. ma fischler, Comm. ACM 24 (6) (1981) 381–395

    1981

  16. [24]

    Hartley, A

    R. Hartley, A. Zisserman, Multiple view geometry in computer vision, Cambridge university press, 2003

    2003

  17. [25]

    Triggs, P

    B. Triggs, P. F. McLauchlan, R. I. Hartley, A. W. Fitzgibbon, Bundle adjustment—a modern synthesis, in: Vision Algorithms: Theory and Practice: International Workshop on Vision Algorithms Corfu, Greece, September 21–22, 1999 Proceedings, Springer, 2000, pp. 298–372

    1999

  18. [26]

    Huang, R

    D. Huang, R. Qin, M. Elhashash, Bundle adjustment with motion con- straints for uncalibrated multi-camera systems at the ground level, IS- PRS Journal of Photogrammetry and Remote Sensing 211 (2024) 452– 464

    2024

  19. [27]

    DeTone, T

    D. DeTone, T. Malisiewicz, A. Rabinovich, Superpoint: Self-supervised interest point detection and description, in: Proceedings of the IEEE conference on computer vision and pattern recognition workshops, 2018, pp. 224–236

    2018

  20. [28]

    Sarlin, D

    P.-E. Sarlin, D. DeTone, T. Malisiewicz, A. Rabinovich, Superglue: Learning feature matching with graph neural networks, in: Proceedings 27 of the IEEE/CVF conference on computer vision and pattern recogni- tion, 2020, pp. 4938–4947

    2020

  21. [29]

    S. Tang, Y. Li, J. Wan, Y. Li, B. Zhou, R. Guo, W. Wang, Y. Feng, Transcnnloc: End-to-end pixel-level learning for 2d-to-3d pose estima- tion in dynamic indoor scenes, ISPRS Journal of Photogrammetry and Remote Sensing 207 (2024) 218–230

    2024

  22. [30]

    Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, B. Guo, Swin transformer: Hierarchical vision transformer using shifted windows, in: Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 10012–10022

    2021

  23. [31]

    C. M. Parameshwara, G. Hari, C. Ferm¨ uller, N. J. Sanket, Y. Aloimonos, Diffposenet: Direct differentiable camera pose estimation, in: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 6845–6854

    2022

  24. [32]

    J. Ho, A. Jain, P. Abbeel, Denoising diffusion probabilistic models, Ad- vances in neural information processing systems 33 (2020) 6840–6851

    2020

  25. [33]

    L. Jin, S. Qian, A. Owens, D. F. Fouhey, Planar surface reconstruc- tion from sparse views, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 12991–13000

    2021

  26. [34]

    J. Y. Zhang, D. Ramanan, S. Tulsiani, Relpose: Predicting probabilistic relative rotation for single objects in the wild, in: European Conference on Computer Vision, Springer, 2022, pp. 592–611

    2022

  27. [35]

    Cerkezi, P

    L. Cerkezi, P. Favaro, Sparse 3d reconstruction via object-centric ray sampling, in: 2024 International Conference on 3D Vision (3DV), IEEE, 2024, pp. 432–441

    2024

  28. [36]

    Umeyama, Least-squares estimation of transformation parameters be- tween two point patterns, IEEE Transactions on Pattern Analysis & Machine Intelligence 13 (04) (1991) 376–380

    S. Umeyama, Least-squares estimation of transformation parameters be- tween two point patterns, IEEE Transactions on Pattern Analysis & Machine Intelligence 13 (04) (1991) 376–380. 28

    1991

  29. [48]

    doi:https://doi.org/10.1016/j.isprsjprs.2024.03.016

    2024 doi

  30. [213]

    doi:https://doi.org/10.1016/j.isprsjprs.2024.10.027

    2024 doi

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.