Pith. sign in

REVIEW 3 major objections 3 minor 60 references

Learning Correlation-aware Aleatoric Uncertainty for 3D Hand Pose Estimation

T0 review · 3 major / 3 minor · reviewed 2026-08-05 · deepseek-v4-flash

Pith's one-line read This paper proposes an uncertainty head for 3D hand pose estimation whose covariance is W diag(σ²)W^T, claiming it captures inter-joint correlations better than diagonal or full-covariance baselines while keeping pose accuracy.

desk verdict The correlation layer's training objective drives it to zero, so the paper's central claim is unsupported despite clean experiments. read the letter →

arxiv 2509.01242 v1 pith:RJOHLIRT submitted 2025-09-01 cs.CV

classification cs.CV
keywords 3Dhandposeestimationaleatoricuncertaintyheteroscedasticjointcorrelationstructuredcovariancenegativelog-likelihoodsparsificationcurvestransformerhead
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 tackles a blind spot in 3D hand pose estimation: models say where joints are but not how sure they are, and standard uncertainty heads treat joints independently even though fingers move together. It introduces aleatoric (data-driven) uncertainty modeling into an existing hand pose transformer and proposes a middle-ground covariance parametrization: a diagonal per-joint variance matrix sandwiched by a shared linear layer, giving W diag(σ²)W^T. This captures joint correlations with far fewer parameters than a full covariance matrix and more expressiveness than a diagonal one. Experiments on two standard benchmarks show the resulting uncertainty scores track actual per-joint errors better than diagonal or full covariance baselines, while 3D pose accuracy is maintained. The practical payoff is an uncertainty signal a downstream system can trust when occlusion, blur, or self-similarity make a hand pose genuinely ambiguous.

What carries the argument

The central object is the low-rank structured covariance W diag(σ²3D)W^T. Per-joint variances σ²3D are produced by an extra transformer head trained with a diagonal Gaussian NLL loss, which defines a zero-mean probabilistic output space p(z|x) = N(0, diag(σ²3D)). Samples z are fed through a single linear layer W; because linear maps preserve Gaussianity, the resulting covariance is exactly W diag(σ²3D)W^T, and the mean µ3D is added afterward. W is shared across inputs and is the mechanism that injects inter-joint correlation into the uncertainty estimate, giving an analytically tractable covariance without sampling-based covariance estimation. This single matrix is what the paper calls the m

What would settle it

Train the proposed head on one of the benchmarks, then for each test image compare the average outer product of residuals y − µ3D with the average predicted covariance W diag(σ²3D)W^T. If the predicted covariance does not approximate the empirical residual covariance—for example, if W collapses toward zero while pose accuracy stays fixed—the claim that the linear layer learns joint correlations is not supported. A second check: fix the mean µ3D and train only the uncertainty terms; if the uncertainty ranking gains vanish, the gains come from improved pose estimates rather than from correlation

Watch

Extended reading notes

Core claim

The paper proposes a new parametrization for aleatoric uncertainty in 3D hand pose estimation. Instead of predicting only per-joint variances or a full covariance matrix, the network regresses per-joint variances under an independence assumption, then passes samples from that zero-mean Gaussian through one learned linear layer W. Because linear maps preserve Gaussianity, the predictive distribution becomes N(µ3D, W diag(σ²3D)W^T). The shared W is the only extra correlation-carrying component, turning the cheap diagonal model into a correlated model with d_o² additional parameters. On FreiHAND and HO3Dv2, the paper reports that this head beats diagonal and full covariance baselines on sparsif

Load-bearing premise

The method assumes that the training loss teaches the linear transformation to encode real correlations between joint errors, rather than simply shrinking toward zero, because nothing in the loss forces the learned covariance to match the actual pattern of prediction errors.

Editorial extensions

If this is right

  • The proposed head mounts on top of an existing pretrained hand pose model, so uncertainty estimation can be added without redesigning the feature extractor or regressor.
  • Correlation-aware uncertainty improves uncertainty quality metrics (AUSC, AUSE, Pearson's ρ) over both diagonal and full covariance heads on FreiHAND and HO3Dv2.
  • Adding the linear layer improves pose accuracy slightly compared to the same head without it, so correlation modeling helps rather than hurts regression.
  • The parametrization requires far fewer parameters than a full covariance matrix, making it practical for the 63-dimensional 3D joint output space.
  • In the 25-sample ablation, more samples consistently improve uncertainty estimates, so performance can be traded against sampling cost.

Reading between the lines

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

  • Inference beyond the paper: if W is trained across many datasets, its off-diagonal entries may encode anatomical hand synergies (e.g., adjacent fingers moving together); this could be tested by clustering the learned W or comparing it to known joint-coupling matrices.
  • Inference beyond the paper: the same W diag(σ²)W^T device should transfer to any structured output with correlated components—body pose, hand-object interaction, facial landmarks—and the paper's efficiency argument would be strongest there.
  • Inference beyond the paper: the reported gains could be stratified by occlusion and blur; if the uncertainty advantage concentrates on occluded or blurred joints, the method's practical case for in-the-wild deployment becomes concrete.
  • Inference beyond the paper: a direct test of whether W diag(σ²)W^T matches the empirical residual covariance (rather than acting mainly as a learned rescaling of per-joint variance) would separate correlation learning from pose-accuracy improvements.
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

3 major / 3 minor

Summary. The paper addresses aleatoric uncertainty estimation for 3D hand pose estimation. The authors propose to model the predictive distribution as N(μ, W diag(σ²) W^T), where μ is the mean hand pose, σ² are per-joint variances estimated by a diagonal uncertainty head, and W is a single learned linear layer intended to capture correlations among hand joints. Training combines the deterministic HaMeR losses, a diagonal negative log-likelihood term (Eq. 2), and an MSE loss on samples drawn from the noise model (Eq. 6). Experiments on FreiHAND and HO3Dv2 report improved uncertainty metrics (AUSC, AUSE, Pearson's ρ) over diagonal and full-covariance baselines, and competitive pose accuracy. The central claim is that the linear layer W learns intrinsic hand joint correlations and provides a favorable efficiency/expressiveness trade-off.

Significance. If the proposed mechanism were sound, the parameterization would be a useful intermediate between diagonal and full covariance modeling: it introduces O(d_o^2) parameters but avoids full covariance's O(d_o^2) per-input cost and optimization instability. The paper also demonstrates a practical application of uncertainty estimation to hand pose estimation. However, the significance is conditional on the training objective actually fitting W to the residual covariance, and the manuscript does not establish that. The gradient analysis below shows that the MSE loss drives W toward zero and no term in the objective rewards W for capturing correlations, so the central claim is not supported. The empirical improvements, while present, are not explained by the proposed mechanism. The paper's internal inconsistency between the claimed predictive distribution and the training loss make the contribution, as presented, unsound.

major comments (3)
  1. [§3.2, Eq. (6)-(7)] The MSE loss does not train W to model joint correlations. Let z ~ N(0, diag(σ²)) be the synthetic noise and r = y − μ. Then L_MSE = E||r − Wz||². Since z is conditionally independent of the data residual r given x, E[r z^T] = 0 and E[z z^T] = diag(σ²), so ∇_W L_MSE = 2 λ_MSE W diag(σ²). This is a contraction toward W = 0; no term in Eq. (7)—neither the diagonal NLL (Eq. 2) nor L_DETER—involves W. Therefore, the training objective provides no mechanism for W to learn the residual covariance. The paper's claim that 'we naturally capture the correlation between hand joints' by this sampling and linear transformation is unsupported. No gradient analysis or control experiment is provided to rule out this collapse.
  2. [§3.2, Eq. (5) vs Eq. (2)] The model defines the predictive covariance as Σ = W diag(σ²) W^T (Eq. 5), but the NLL term used in training is the diagonal NLL with covariance diag(σ²) (Eq. 2). The full-covariance NLL (Eq. 4) is used only for the Full baseline, not for the proposed model. Consequently, the training objective does not correspond to the negative log-likelihood of the distribution p(y|x) = N(μ, W diag(σ²) W^T) that the paper claims to model. The learned σ² and W are therefore not maximum-likelihood estimates of the proposed probabilistic model, and the uncertainty reported at inference time is not the quantity optimized during training. This internal inconsistency undermines the principled basis of the method.
  3. [§4.4, Table 4; §4.2, Table 2] The evidence that W captures 'intrinsic hand joint correlations' is self-referential. The only support is the improvement in AUSC/AUSE/Pearson on the same benchmarks. No analysis of the learned W is presented—for example, no visualization of the implied correlation matrix W W^T, no comparison to empirical residual covariances, and no comparison to anatomical priors. The ablation 'Ours w/o linear layer' does not control for W initialization or training dynamics. Given the gradient analysis above, the reported improvements could stem from the stochastic noise injection acting as a regularizer on the variance head, or from a near-identity W that is slowly shrinking, rather than from learned correlations. A control with W fixed to identity (or to a random constant) under the same training pipeline is needed to attribute the improvements to correlation learning.
minor comments (3)
  1. [Eq. (2) and Eq. (4)] The equations are written as L_NLL = log p(...), but the text says these are negative log-likelihood losses. Without a minus sign, minimizing L_NLL would maximize the likelihood, which is the opposite of the stated intent. Please add the minus sign or a note that the constant is absorbed.
  2. [References [26] and [27]] References [26] and [27] cite the same work (Kendall and Gal, 2017) in two different formats. These duplicates should be merged.
  3. [Eq. (6)] The expectation in Eq. (6) is written without a subscript. It should be E_z or E_{z~N(0,diag(σ²))} to make clear that the expectation is over the sampled noise.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the covariance parametrization is a valid construction, and the training-objective mismatch is a correctness risk rather than a circular reduction.

full rationale

The paper's central construction is self-contained: given p(z|x) ~ N(0, diag(sigma^2_3D)) and y = mu_3D + W z, linearity of Gaussians implies p(y|x) ~ N(mu_3D, W diag(sigma^2_3D) W^T) (Eq. 5). This is a mathematical identity of the reparametrization, not a circular prediction. The diagonal NLL (Eq. 2), MSE (Eq. 6), and total loss (Eq. 7) are stated explicitly, and the empirical evaluation compares against external FreiHAND and HO3Dv2 benchmarks and two baselines, with an external competing method (AMVUR) also evaluated. There is no load-bearing self-citation: the pretrained backbone HaMeR [42] is an external work, and no 'uniqueness' or 'ansatz' is imported from the authors' prior publications. The skeptic's concern is real but is an optimization/validity gap, not circularity: because z is independent of the residual (y - mu), the MSE term's W-gradient is 2 W diag(sigma^2), driving W toward zero, and the diagonal NLL in Eq. (2) does not involve W; thus nothing in Eq. (7) forces W to match the true residual covariance. This means the claim that 'W learns joint correlations' is underjustified, and the trace-based AUSC metric does not directly measure off-diagonal correlation. However, underjustification and objective mismatch do not make the derivation equivalent to its inputs by construction, so per the circularity definitions in scope, no circular step is present.

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

The central covariance depends on hand-set weights lambda_NLL, lambda_MSE, sample count N, and the learned matrix W. The paper provides no sensitivity analysis for the loss weights and no analysis of W's convergence, especially given that the MSE objective drives W toward zero.

free parameters (4)
  • Loss weight lambda_NLL = 5e-4
    Set in Sec. 4.1 implementation details; no sensitivity analysis reported.
  • Loss weight lambda_MSE = 5e-4
    Set in Sec. 4.1 implementation details; controls the sampling MSE term in Eq. (7), which is the only term that updates W.
  • Number of samples N = 25
    Ablation in Table 7 shows 25 is chosen as a performance/compute trade-off.
  • Linear layer W (63x63) = learned, not reported
    Central to the covariance Sigma = W diag(sigma^2) W^T; its initialization and final values are not given, and the stated objective drives it toward zero.
assumptions (4)
  • domain assumption Hand joint output follows a Gaussian distribution in the proposed output space
    Sec. 3.2 defines p(z|x) ~ N(0, diag(sigma^2)) and p(y|z) ~ N(mu, W diag(sigma^2) W^T); no evidence that hand joint errors are Gaussian.
  • standard math Linearity of Gaussian under affine maps
    Used to derive Eq. (5); standard and safe.
  • domain assumption Aleatoric uncertainty is heteroscedastic and input-dependent
    Sec. 2, following Kendall and Gal; assumed rather than tested on hand pose data.
  • ad hoc to paper A single fixed W represents intrinsic hand joint correlations
    Sec. 3.2 and Table 1; the paper assumes one shared matrix captures joint correlations across all inputs, but does not validate against known anatomical correlations or show that W is identifiable under the training objective.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning Correlation-aware Aleatoric Uncertainty for 3D Hand Pose Estimation." pith.science (2026). https://pith.science/paper/RJOHLIRT

@misc{pith2026250901242,
  author       = {Pith},
  title        = {Pith review of: Learning Correlation-aware Aleatoric Uncertainty for 3D Hand Pose Estimation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RJOHLIRT}},
  note         = {Machine review of arXiv:2509.01242}
}
read the original abstract

3D hand pose estimation is a fundamental task in understanding human hands. However, accurately estimating 3D hand poses remains challenging due to the complex movement of hands, self-similarity, and frequent occlusions. In this work, we address two limitations: the inability of existing 3D hand pose estimation methods to estimate aleatoric (data) uncertainty, and the lack of uncertainty modeling that incorporates joint correlation knowledge, which has not been thoroughly investigated. To this end, we introduce aleatoric uncertainty modeling into the 3D hand pose estimation framework, aiming to achieve a better trade-off between modeling joint correlations and computational efficiency. We propose a novel parameterization that leverages a single linear layer to capture intrinsic correlations among hand joints. This is enabled by formulating the hand joint output space as a probabilistic distribution, allowing the linear layer to capture joint correlations. Our proposed parameterization is used as a task head layer, and can be applied as an add-on module on top of the existing models. Our experiments demonstrate that our parameterization for uncertainty modeling outperforms existing approaches. Furthermore, the 3D hand pose estimation model equipped with our uncertainty head achieves favorable accuracy in 3D hand pose estimation while introducing new uncertainty modeling capability to the model. The project page is available at https://hand-uncertainty.github.io/.

Figures

Figures reproduced from arXiv: 2509.01242 by the authors.

Figure 1
Figure 1. Illustration of deterministic modeling and correlation modelings of uncertainty. (a) Deterministic modeling produces a single deterministic output, represented as point embedding in the output space; (b) Diagonal and (c) Full correlation modeling of output produce means and covariances of a Gaussian distribution, where the uncertainty of the prediction is modeled by its variance. (d) Ours learns the variation of eac… view at source ↗
Figure 2
Figure 2. Pipeline of our proposed method. We train a pre-trained large model [42] by introducing an additional transformer head that estimates the variance for each joint under the independent assumption. This estimated uncertainty then defines a probabilistic hand joint output space, from which we sample and pass the samples through a single linear layer to model the correlations between hand joints. 3.2 Correlation-Aware A… view at source ↗
Figure 3
Figure 3. Sparsification curves. We compare sparsification curves obtained by different methods of es￾timating the uncertainty of 3D hand joints. Metrics on 3D hand pose estimation. We follow the typical protocols used in previous works [32, 40, 42], and report PA-MPJPE and AUCJ for evaluating estimated 3D hand joints and PA-MPVPE, AUCV, F@5mm and F@15mm for evaluating estimated 3D hand mesh. PA-MPJPE and PA-MPVPE are measure… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Qualitative results of uncertainty estimation. We evaluate the quality of un￾certainty estimates by comparing our method with existing uncertainty modeling methods. Specifically, we visualize the prediction errors alongside the corresponding uncertainty values. A desir…
Figure 5
Figure 5. Figure 5: 2D histograms of model error and estimated uncertainty. We show 2D his￾tograms of the x-axis representing the model’s MPJPE error and the y-axis representing the estimated uncertainty on (a) FreiHAND [59] and (b) HO3Dv2 [18] datasets. Implementation details. In all exp…
Figure 6
Figure 6. Figure 6: Sparsification curve of AMVUR [24]. The competing method exhibits a large discrepancy between uncertainty-based and error-based sortings. Method AUSC ↓ AUSE ↓ Pearson’s ρ ↑ AMVUR [24] 586 113 0.127 Ours 505 57.6 0.600 [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 52 canonical work pages

  1. [1]

    Pushing the envelope for rgb-based dense 3d hand pose estimation via neural rendering

    Seungryul Baek, Kwang In Kim, and Tae-Kyun Kim. Pushing the envelope for rgb-based dense 3d hand pose estimation via neural rendering. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 1067–1076, 2019

  2. [2]

    Vision- based hand posture detection and recognition for sign language—a study

    Sara Bilal, Rini Akmeliawati, Momoh Jimoh El Salami, and Amir A Shafie. Vision- based hand posture detection and recognition for sign language—a study. In 2011 4th International Conference on Mechatronics (ICOM), pages 1–6. IEEE, 2011

  3. [3]

    Plausible uncertainties for human pose regression

    Lennart Bramlage, Michelle Karg, and Cristóbal Curio. Plausible uncertainties for human pose regression. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 15133–15142, 2023

  4. [4]

    Weakly-supervised 3d hand pose estimation from monocular rgb images

    Yujun Cai, Liuhao Ge, Jianfei Cai, and Junsong Yuan. Weakly-supervised 3d hand pose estimation from monocular rgb images. In Proceedings of the European conference on computer vision (ECCV), pages 666–682, 2018

  5. [5]

    Active learning for bayesian 3d hand pose estimation

    Razvan Caramalau, Binod Bhattarai, and Tae-Kyun Kim. Active learning for bayesian 3d hand pose estimation. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 3419–3428, 2021

  6. [6]

    Dexycb: A benchmark for capturing hand grasping of objects

    Yu-Wei Chao, Wei Yang, Yu Xiang, Pavlo Molchanov, Ankur Handa, Jonathan Tremblay, Yashraj S Narang, Karl Van Wyk, Umar Iqbal, Stan Birchfield, et al. Dexycb: A benchmark for capturing hand grasping of objects. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 9044–9053, 2021

  7. [7]

    I2uv-handnet: Image-to-uv prediction network for accurate and high-fidelity 3d hand mesh modeling

    Ping Chen, Yujin Chen, Dong Yang, Fangyin Wu, Qin Li, Qingpei Xia, and Yong Tan. I2uv-handnet: Image-to-uv prediction network for accurate and high-fidelity 3d hand mesh modeling. In Proceedings of the IEEE/CVF international conference on computer vision, pages 12929–12938, 2021

  8. [8]

    Mobrecon: Mobile-friendly hand mesh reconstruction from monocular image

    Xingyu Chen, Yufeng Liu, Yajiao Dong, Xiong Zhang, Chongyang Ma, Yanmin Xiong, Yuan Zhang, and Xiaoyan Guo. Mobrecon: Mobile-friendly hand mesh reconstruction from monocular image. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 20544–20554, 2022. 12 CHAE-YEON ET AL.: LEARNING CORRELA TION-AW ARE ALEA TORIC UNCERTAINTY

Show all 60 references
  1. [9]

    Estimating ego-body pose from doubly sparse egocentric video data

    Seunggeun Chi, Pin-Hao Huang, Enna Sachdeva, Hengbo Ma, Karthik Ramani, and Kwonjoon Lee. Estimating ego-body pose from doubly sparse egocentric video data. arXiv preprint arXiv:2411.03561, 2024

  2. [10]

    Pose2mesh: Graph convo- lutional network for 3d human pose and mesh recovery from a 2d human pose

    Hongsuk Choi, Gyeongsik Moon, and Kyoung Mu Lee. Pose2mesh: Graph convo- lutional network for 3d human pose and mesh recovery from a 2d human pose. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part VII 16, pages 769–787....

  3. [11]

    Aleatory or epistemic? does it matter? Structural safety, 31(2):105–112, 2009

    Armen Der Kiureghian and Ove Ditlevsen. Aleatory or epistemic? does it matter? Structural safety, 31(2):105–112, 2009

  4. [12]

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

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv...

  5. [13]

    Multi-finger grasping like humans

    Yuming Du, Philippe Weinzaepfel, Vincent Lepetit, and Romain Brégier. Multi-finger grasping like humans. In 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1564–1570. IEEE, 2022

  6. [14]

    Arctic: A dataset for dexterous bimanual hand- object manipulation

    Zicong Fan, Omid Taheri, Dimitrios Tzionas, Muhammed Kocabas, Manuel Kaufmann, Michael J Black, and Otmar Hilliges. Arctic: A dataset for dexterous bimanual hand- object manipulation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1...

  7. [15]

    Alphapose: Whole-body regional multi-person pose estimation and tracking in real-time

    Hao-Shu Fang, Jiefeng Li, Hongyang Tang, Chao Xu, Haoyi Zhu, Yuliang Xiu, Yong-Lu Li, and Cewu Lu. Alphapose: Whole-body regional multi-person pose estimation and tracking in real-time. IEEE transactions on pattern analysis and machine intelligence , 45(6):7157–7173, 2022

  8. [16]

    3d hand shape and pose estimation from a single rgb image

    Liuhao Ge, Zhou Ren, Yuncheng Li, Zehao Xue, Yingying Wang, Jianfei Cai, and Junsong Yuan. 3d hand shape and pose estimation from a single rgb image. In Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition , pages 10833–10842, 2019

  9. [17]

    Quantifying the independence of human finger movements: comparisons of digits, hands, and movement frequencies

    Charlotte Häger-Ross and Marc H Schieber. Quantifying the independence of human finger movements: comparisons of digits, hands, and movement frequencies. Journal of Neuroscience, 20(22):8542–8550, 2000

  10. [18]

    Honnotate: A method for 3d annotation of hand and object poses

    Shreyas Hampali, Mahdi Rad, Markus Oberweger, and Vincent Lepetit. Honnotate: A method for 3d annotation of hand and object poses. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 3196–3206, 2020

  11. [19]

    Dexpilot: Vision-based teleoperation of dexterous robotic hand-arm system

    Ankur Handa, Karl Van Wyk, Wei Yang, Jacky Liang, Yu-Wei Chao, Qian Wan, Stan Birchfield, Nathan Ratliff, and Dieter Fox. Dexpilot: Vision-based teleoperation of dexterous robotic hand-arm system. In 2020 IEEE International Conference on Robotics and Automation (ICRA), pages 9...

  12. [20]

    Learning joint reconstruction of hands and manipulated objects

    Yana Hasson, Gul Varol, Dimitrios Tzionas, Igor Kalevatykh, Michael J Black, Ivan Laptev, and Cordelia Schmid. Learning joint reconstruction of hands and manipulated objects. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11807–1181...

  13. [21]

    A quantitative evaluation of confidence measures for stereo vision

    Xiaoyan Hu and Philippos Mordohai. A quantitative evaluation of confidence measures for stereo vision. IEEE transactions on pattern analysis and machine intelligence , 34 (11):2121–2133, 2012

  14. [22]

    Uncertainty estimates and multi-hypotheses networks for optical flow

    Eddy Ilg, Ozgun Cicek, Silvio Galesso, Aaron Klein, Osama Makansi, Frank Hutter, and Thomas Brox. Uncertainty estimates and multi-hypotheses networks for optical flow. In Proceedings of the European Conference on Computer Vision (ECCV) , pages 652–667, 2018

  15. [23]

    The statistics of natural hand movements

    James N Ingram, Konrad P Körding, Ian S Howard, and Daniel M Wolpert. The statistics of natural hand movements. Experimental brain research, 188:223–236, 2008

  16. [24]

    A probabilistic attention model with occlusion-aware texture regression for 3d hand reconstruction from a single rgb image

    Zheheng Jiang, Hossein Rahmani, Sue Black, and Bryan M Williams. A probabilistic attention model with occlusion-aware texture regression for 3d hand reconstruction from a single rgb image. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pa...

  17. [25]

    Whole-body human pose estimation in the wild

    Sheng Jin, Lumin Xu, Jin Xu, Can Wang, Wentao Liu, Chen Qian, Wanli Ouyang, and Ping Luo. Whole-body human pose estimation in the wild. In European Conference on Computer Vision, pages 196–214. Springer, 2020

  18. [26]

    What uncertainties do we need in bayesian deep learning for computer vision? Advances in neural information processing systems , 30, 2017

    Alex Kendall and Yarin Gal. What uncertainties do we need in bayesian deep learning for computer vision? Advances in neural information processing systems , 30, 2017

  19. [27]

    What uncertainties do we need in bayesian deep learning for computer vision? In NIPS, 2017

    Alex Kendall and Yarin Gal. What uncertainties do we need in bayesian deep learning for computer vision? In NIPS, 2017

  20. [28]

    End-to-end detection and pose estimation of two interacting hands

    Dong Uk Kim, Kwang In Kim, and Seungryul Baek. End-to-end detection and pose estimation of two interacting hands. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 11189–11198, 2021

  21. [29]

    Weakly-supervised mesh-convolutional hand reconstruction in the wild

    Dominik Kulon, Riza Alp Guler, Iasonas Kokkinos, Michael M Bronstein, and Stefanos Zafeiriou. Weakly-supervised mesh-convolutional hand reconstruction in the wild. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 4990–5000, 2020

  22. [30]

    Interacting attention graph for single image two-hand reconstruction

    Mengcheng Li, Liang An, Hongwen Zhang, Lianpeng Wu, Feng Chen, Tao Yu, and Yebin Liu. Interacting attention graph for single image two-hand reconstruction. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 2761–2770, 2022

  23. [31]

    End-to-end human pose and mesh recon- struction with transformers

    Kevin Lin, Lijuan Wang, and Zicheng Liu. End-to-end human pose and mesh recon- struction with transformers. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1954–1963, 2021

  24. [32]

    Mesh graphormer

    Kevin Lin, Lijuan Wang, and Zicheng Liu. Mesh graphormer. In Proceedings of the IEEE/CVF international conference on computer vision , pages 12939–12948, 2021

  25. [33]

    Analysis of hand and wrist postural synergies in tolerance grasping of various objects

    Yuan Liu, Li Jiang, Dapeng Yang, and Hong Liu. Analysis of hand and wrist postural synergies in tolerance grasping of various objects. PloS one, 11(8):e0161772, 2016. 14 CHAE-YEON ET AL.: LEARNING CORRELA TION-AW ARE ALEA TORIC UNCERTAINTY

  26. [34]

    Quantitative investigation of hand grasp functionality: hand joint motion correlation, independence, and grasping behavior

    Yuan Liu, Bo Zeng, Ting Zhang, Li Jiang, Hong Liu, and Dong Ming. Quantitative investigation of hand grasp functionality: hand joint motion correlation, independence, and grasping behavior. Applied Bionics and Biomechanics , 2021(1):2787832, 2021

  27. [35]

    3d interacting hand pose estimation by hand de-occlusion and removal

    Hao Meng, Sheng Jin, Wentao Liu, Chen Qian, Mengxiang Lin, Wanli Ouyang, and Ping Luo. 3d interacting hand pose estimation by hand de-occlusion and removal. In European Conference on Computer Vision, pages 380–397. Springer, 2022

  28. [36]

    I2l-meshnet: Image-to-lixel prediction network for accurate 3d human pose and mesh estimation from a single rgb image

    Gyeongsik Moon and Kyoung Mu Lee. I2l-meshnet: Image-to-lixel prediction network for accurate 3d human pose and mesh estimation from a single rgb image. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part VII 16, pages 752...

  29. [37]

    Inter- hand2

    Gyeongsik Moon, Shoou-I Yu, He Wen, Takaaki Shiratori, and Kyoung Mu Lee. Inter- hand2. 6m: A dataset and baseline for 3d interacting hand pose estimation from a single rgb image. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedi...

  30. [38]

    Gallagher, Kevin P

    Seong Joon Oh, Andrew C. Gallagher, Kevin P. Murphy, Florian Schroff, Jiyan Pan, and Joseph Roth. Modeling uncertainty with hedged instance embeddings. In ICLR, 2019

  31. [39]

    Recovering 3d hand mesh sequence from a single blurry image: A new dataset and temporal unfolding

    Yeonguk Oh, JoonKyu Park, Jaeha Kim, Gyeongsik Moon, and Kyoung Mu Lee. Recovering 3d hand mesh sequence from a single blurry image: A new dataset and temporal unfolding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 554–563, 2023

  32. [40]

    Handoccnet: Occlusion-robust 3d hand mesh estimation network

    JoonKyu Park, Yeonguk Oh, Gyeongsik Moon, Hongsuk Choi, and Kyoung Mu Lee. Handoccnet: Occlusion-robust 3d hand mesh estimation network. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 1496–1505, 2022

  33. [41]

    3d hand sequence recovery from real blurry images and event stream

    Joonkyu Park, Gyeongsik Moon, Weipeng Xu, Evan Kaseman, Takaaki Shiratori, and Kyoung Mu Lee. 3d hand sequence recovery from real blurry images and event stream. In European Conference on Computer Vision, pages 343–359. Springer, 2024

  34. [42]

    Reconstructing hands in 3d with transformers

    Georgios Pavlakos, Dandan Shan, Ilija Radosavovic, Angjoo Kanazawa, David Fouhey, and Jitendra Malik. Reconstructing hands in 3d with transformers. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 9826–9836, 2024

  35. [43]

    On the uncertainty of self-supervised monocular depth estimation

    Matteo Poggi, Filippo Aleotti, Fabio Tosi, and Stefano Mattoccia. On the uncertainty of self-supervised monocular depth estimation. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 3227–3237, 2020

  36. [44]

    Dexmv: Imitation learning for dexterous manipulation from human videos

    Yuzhe Qin, Yueh-Hua Wu, Shaowei Liu, Hanwen Jiang, Ruihan Yang, Yang Fu, and Xiaolong Wang. Dexmv: Imitation learning for dexterous manipulation from human videos. In European Conference on Computer Vision, pages 570–587. Springer, 2022

  37. [45]

    Embodied hands: Modeling and capturing hands and bodies together

    Javier Romero, Dimitrios Tzionas, and Michael J Black. Embodied hands: Modeling and capturing hands and bodies together. arXiv preprint arXiv:2201.02610, 2022. CHAE-YEON ET AL.: LEARNING CORRELA TION-AW ARE ALEA TORIC UNCERTAINTY 15

  38. [46]

    Frankmocap: A monocular 3d whole- body pose estimation system via regression and integration

    Yu Rong, Takaaki Shiratori, and Hanbyul Joo. Frankmocap: A monocular 3d whole- body pose estimation system via regression and integration. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 1749–1759, 2021

  39. [47]

    Muscular production of individuated finger movements: the roles of extrinsic finger muscles

    Marc H Schieber. Muscular production of individuated finger movements: the roles of extrinsic finger muscles. Journal of Neuroscience, 15(1):284–297, 1995

  40. [48]

    Hand keypoint detection in single images using multiview bootstrapping

    Tomas Simon, Hanbyul Joo, Iain Matthews, and Yaser Sheikh. Hand keypoint detection in single images using multiview bootstrapping. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, pages 1145–1153, 2017

  41. [49]

    Constraining dense hand surface tracking with elasticity

    Breannan Smith, Chenglei Wu, He Wen, Patrick Peluse, Yaser Sheikh, Jessica K Hod- gins, and Takaaki Shiratori. Constraining dense hand surface tracking with elasticity. ACM Transactions on Graphics (ToG), 39(6):1–14, 2020

  42. [50]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017

  43. [51]

    Handflow: Quantifying view-dependent 3d ambigu- ity in two-hand reconstruction with normalizing flow

    Jiayi Wang, Diogo Luvizon, Franziska Mueller, Florian Bernard, Adam Kortylewski, Dan Casas, and Christian Theobalt. Handflow: Quantifying view-dependent 3d ambigu- ity in two-hand reconstruction with normalizing flow. arXiv preprint arXiv:2210.01692, 2022

  44. [52]

    Monocular total capture: Posing face, body, and hands in the wild

    Donglai Xiang, Hanbyul Joo, and Yaser Sheikh. Monocular total capture: Posing face, body, and hands in the wild. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10965–10974, 2019

  45. [53]

    Artiboost: Boosting articulated 3d hand-object pose estimation via online exploration and synthesis

    Lixin Yang, Kailin Li, Xinyu Zhan, Jun Lv, Wenqiang Xu, Jiefeng Li, and Cewu Lu. Artiboost: Boosting articulated 3d hand-object pose estimation via online exploration and synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 275...

  46. [54]

    Acr: Attention collaboration-based regressor for arbitrary two-hand reconstruction

    Zhengdi Yu, Shaoli Huang, Chen Fang, Toby P Breckon, and Jue Wang. Acr: Attention collaboration-based regressor for arbitrary two-hand reconstruction. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 12955– 12964, 2023

  47. [55]

    Interacting two-hand 3d pose and shape reconstruction from single color image

    Baowen Zhang, Yangang Wang, Xiaoming Deng, Yinda Zhang, Ping Tan, Cuixia Ma, and Hongan Wang. Interacting two-hand 3d pose and shape reconstruction from single color image. In Proceedings of the IEEE/CVF international conference on computer vision, pages 11354–11363, 2021

  48. [56]

    Heteroscedastic uncer- tainty estimation framework for unsupervised registration

    Xiaoran Zhang, Daniel H Pak, Shawn S Ahn, Xiaoxiao Li, Chenyu You, Lawrence H Staib, Albert J Sinusas, Alex Wong, and James S Duncan. Heteroscedastic uncer- tainty estimation framework for unsupervised registration. In International Conference on Medical Image Computing and Co...

  49. [57]

    Weakly-supervised 3d hand reconstruction with knowledge prior and uncertainty guidance

    Yufei Zhang, Jeffrey O Kephart, and Qiang Ji. Weakly-supervised 3d hand reconstruction with knowledge prior and uncertainty guidance. In European Conference on Computer Vision, pages 106–125. Springer, 2024. 16 CHAE-YEON ET AL.: LEARNING CORRELA TION-AW ARE ALEA TORIC UNCERTAINTY

  50. [58]

    Learning to estimate 3d hand pose from single rgb images

    Christian Zimmermann and Thomas Brox. Learning to estimate 3d hand pose from single rgb images. In Proceedings of the IEEE international conference on computer vision, pages 4903–4911, 2017

  51. [59]

    Freihand: A dataset for markerless capture of hand pose and shape from single rgb images

    Christian Zimmermann, Duygu Ceylan, Jimei Yang, Bryan Russell, Max Argus, and Thomas Brox. Freihand: A dataset for markerless capture of hand pose and shape from single rgb images. In Proceedings of the IEEE/CVF international conference on computer vision, pages 813–822, 2019

  52. [60]

    Reconstructing interacting hands with interaction prior from monocular images

    Binghui Zuo, Zimeng Zhao, Wenqian Sun, Wei Xie, Zhou Xue, and Yangang Wang. Reconstructing interacting hands with interaction prior from monocular images. In Proceedings of the IEEE/CVF international conference on computer vision , pages 9054–9064, 2023

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.