Pith. sign in

REVIEW 3 major objections 4 minor 81 references

2D-3D Attention and Entropy for Pose Robust 2D Facial Recognition

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read By training a 2D face network with a shared 2D-3D attention mapping and a joint entropy loss, the paper reports profile-view (90°+) TAR@1%FAR gains of at least 7.1 points on FaceScape and 1.57 points on ARL-VTF while keeping inference…

desk verdict A useful empirical recipe undercut by a mis-specified joint entropy loss; the paper's central mechanism does not do what it claims, though the experiments are solid enough to warrant a serious look if the authors can fix the math. read the letter →

arxiv 2505.09073 v1 pith:H22Q3BZ2 submitted 2025-05-14 cs.CV

J. Brennan Peace , Shuowen Hu , Benjamin S. Riggan This is my paper · ORCID
classification cs.CV
keywords 2D-3Ddomainadaptationpose-invariantfacerecognitionjointattentionmappingentropyregularizationpointcloudpriorsprofileTAR@1%FAR
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 aims to make ordinary 2D face recognition hold up under large pose differences by letting the 2D network, during training only, learn from pose-invariant 3D point-cloud representations. Its two new pieces are a shared 2D-3D joint attention mapping (JAM), which computes attention for both modalities with the same filters, and a joint entropy (JE) loss, which is intended to keep the two attention maps consistent. Evaluated on FaceScape, the combined framework reaches 78.910% TAR@1%FAR at 90°+ profile pose, at least 7.1 percentage points above the compared domain-adaptation baselines; on the 2D-only ARL-VTF benchmark it improves profile TAR@1%FAR by 1.57 points over the best comparison. If the result holds, deployed face recognition could gain pose invariance without needing 3D sensors, 3D enrollment data, or frontalization at inference time.

What carries the argument

The load-bearing mechanism is the pair (JAM, JE). JAM is the attention operation $J_d = \gamma\,\mathrm{softmax}(Q(z_d)K^\top(z_d))V_d(z_d) + z_d$, where $Q$ and $K$ are 1x1 convolutions shared between the 2D and 3D branches, so both modalities are forced to attend with the same filters, while the value projections $V_d$ remain modality-specific. JE is the joint entropy $H(A_{2D}, A_{3D})$ of the discretized attention maps, computed from a normalized histogram over pairs of attention values and minimized together with the two identification losses. The paper's stated mechanism is that minimizing this entropy regularizes the shared attention so that consistent patterns across the two maps are emphasized and pose-variant 2D patterns are de-emphasized, which is what the ablation isolates when JE is removed.

What would settle it

On a trained model, compute the paper's $L_{JE}(A_{2D}, A_{3D})$ and compare it to $H(A_{2D}) + H(A_{3D})$; equality for every batch would show the joint entropy carries no location-pairing information, and any pose-invariance gain would then have to come from the shared attention mapping alone. A direct check is to retrain with the JE term replaced by that marginal-entropy sum and see whether the ARL-VTF profile TAR@1%FAR gain disappears.

Watch

Extended reading notes

Core claim

The central claim is that pose invariance can be transferred from a 3D point-cloud network to a 2D face embedding network through a shared attention mechanism, without ever requiring 3D input at inference. The framework passes 2D and 3D feature maps through identical query and key projections, so the resulting attention maps emphasize the same spatial and channel patterns in both domains, and it adds a joint entropy regularizer computed on the discretized attention maps to penalize inconsistency between them. With this design, the paper reports TAR@1%FAR of 78.910% for 90°+ views on FaceScape, a gap of only 18.19 percentage points from frontal performance, and a profile TAR@1%FAR of 54.678% on ARL-VTF after training only on FaceScape, which beats the compared methods despite a 2.605-point drop on frontal ARL-VTF views that the paper attributes to domain shift.

Load-bearing premise

The load-bearing premise is that the joint entropy of the two attention maps, as built from a histogram over all pairs of discretized attention values, actually couples corresponding 2D and 3D attention locations; if that histogram instead treats the two maps as independent, the regularizer cannot enforce cross-modal consistency by itself.

Editorial extensions

If this is right

  • Deployed 2D face recognition gains pose invariance without any 3D capture at inference time, because 3D data is used only during training.
  • Models trained on a paired 2D-3D dataset transfer to 2D-only operational benchmarks: the framework's ARL-VTF profile TAR@1%FAR is 54.678%, above every compared method, with average TAR@1%FAR of 68.550%.
  • The frontal-to-profile performance gap narrows to 18.19 percentage points on FaceScape, versus 23.78 for the strongest comparison, meaning the benefit is concentrated at the extreme poses that break conventional systems.
  • Because inference needs only a 2D image, the approach can be inserted into existing recognition pipelines without keypoint detection, frontalization, or synthetic-view generation.
  • The main practical limit is the stated requirement for an aligned 2D-3D training corpus; the method inherits the cost and availability constraints of such datasets.

Reading between the lines

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

  • Editorial extension: if the histogram in Eq. 6 sums over all location pairs, the joint entropy equals the sum of the two marginal entropies, so the regularizer would encourage each attention map to be concentrated rather than aligning corresponding 2D and 3D locations; this can be tested by replacing $L_{JE}$ with $H(A_{2D})+H(A_{3D})$ and re-running the reported protocol.
  • Editorial extension: a location-paired loss such as mutual information between corresponding attention positions would directly implement the paper's consistency story and could show whether location-wise alignment, rather than sparsity, drives the profile-view gains.
  • Editorial extension: the recipe of shared attention projections plus an entropy regularizer could generalize to other pose-sensitive 2D tasks that have a 3D morphable model, such as head-pose estimation or expression recognition, although the paper does not claim this.
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

3 major / 4 minor

Summary. The paper proposes a domain-adaptive framework for pose-robust 2D face recognition. It combines a shared 2D-3D Joint Attention Mapping (JAM), which uses shared query/key convolutions to compute attention for both 2D images and 3D point clouds, with a Joint Entropy (JE) regularizing loss that is claimed to promote consistency between the 2D and 3D attention maps. The framework is trained on FaceScape, which provides paired 2D and 3D data, and evaluated on FaceScape and the 2D-only ARL-VTF dataset. The authors report TAR@1% FAR improvements of at least 7.1% and 1.57% for extreme profile poses on FaceScape and ARL-VTF, respectively, and present ablations showing that JAM and JE contribute to the results.

Significance. If the stated mechanism were correct, the paper would offer a practically useful way to inject 3D pose-invariance into 2D face recognition without requiring 3D data at inference. The cross-dataset evaluation on ARL-VTF is a genuine strength, and the JAM idea of computing attention with shared 2D/3D parameters is interesting. However, the central theoretical claim is not supported by the paper's own mathematics: the joint entropy loss defined in Eq. (6) factorizes into the sum of two marginal entropies and therefore cannot enforce any correspondence between 2D and 3D attention locations. This invalidates the abstract's causal story and the motivation in Sections III and IV. The empirical gains in Tables I and III could still stem from JAM plus a low-entropy regularizer, but the paper as written does not establish the claimed cross-modal consistency mechanism. No code or binning details are provided, making the loss irreproducible as defined.

major comments (3)
  1. [Section IV-E, Eq. (6)] The joint distribution P(a2D,a3D) is defined as a normalized sum over all pairs of positions (i,j), i.e., P(a2D,a3D) = (1/N^2) * sum_{i,j} I{A2D(i)=a2D} I{A3D(j)=a3D}. This factorizes exactly as P(a2D) * P(a3D), where P(a2D) = (1/N) sum_i I{A2D(i)=a2D} and similarly for P(a3D). Therefore H(A2D,A3D) = H(A2D) + H(A3D), and minimizing LJE is equivalent to independently minimizing the marginal entropies of the two attention maps. It does not penalize differences or inconsistencies between corresponding 2D and 3D attention locations. The claims in the Abstract, Section III, and Section IV-E that JE 'penalizes differences or inconsistencies between the feature maps' and 'promotes consistency among intersecting 2D and 3D representations' are thus not supported by the equation as written. This is a load-bearing issue because the paper attributes its pose-invariance gains to this loss.
  2. [Section V-C / Section V-D, Tables I and III] All quantitative results are reported as point estimates averaged over 'three independent training cycles,' but no standard deviations, per-fold values, or significance tests are provided. The headline ARL-VTF pose improvement over CrossPoint is 1.571 percentage points (54.678 vs. 53.107), which is small enough that run-to-run variation could change the conclusion. The authors should report the variance across folds and, ideally, a paired significance test, to support the claim that the improvement is systematic rather than noise.
  3. [Section IV-E and Section V-C] The paper does not specify how the joint histogram used in Eq. (6) is computed in practice: the number of bins, the bin boundaries, and whether the sum runs over all (i,j) pairs (as written) or over corresponding positions (A2D(i), A3D(i)) are all unspecified. If the implementation actually used paired indices, then the loss is different from Eq. (6) and the correct formula must be given. As written, the loss cannot be reproduced, and the discrepancy between the equation and any plausible implementation is material.
minor comments (4)
  1. [Section V-E, first sentence] The word 'pase' should be 'pose.'
  2. [Table III] The entry 'Posteret al.' is missing a space, and the footnote marker '1The method...' is formatted inconsistently, making the table hard to read.
  3. [Section V-A] The notation 'G VB0' and 'P VB0' for ARL-VTF gallery/probe designations is not defined in the text or in a reference to the dataset's protocol, which reduces reproducibility.
  4. [Section V-D] The text says 'Without the Joint Entropy minimization, Ours + JAM showcased the highest performance ... on FaceScape,' but Table I shows that Ours+JAM+JE has a higher 90+ TAR@1% FAR (78.910 vs. 78.501). This should be stated more precisely to avoid an apparent contradiction.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation found; the reported gains rest on external benchmarks. The joint entropy loss has a mathematical soundness flaw (it factorizes into marginal entropies), but that is a correctness issue, not circularity.

full rationale

The paper's central performance claims are empirical comparisons on FaceScape and ARL-VTF against external baselines (CrossPoint, AdaFace, MMD, CORAL, etc.), with training and evaluation on disjoint identity splits. The AdaFace margin parameters were tuned on validation data, which is standard hyperparameter selection and does not constitute a fitted input being renamed as a prediction. The self-citations in the reference list provide datasets and baselines (e.g., ARL-VTF, RST), but the core 3D-prior mechanism is built on FaceScape 3DMMs and PointNet++, so those self-citations are not load-bearing for the claimed contribution. The one notable internal issue is mathematical rather than circular: Eq. 6 defines P(a2D,a3D) = (1/N^2) * sum_{i,j} I{A2D(i)=a2D} I{A3D(j)=a3D}, which factorizes into the product of the marginal histograms of A2D and A3D. Consequently H(A2D,A3D) = H(A2D) + H(A3D), and minimizing LJE cannot by itself couple corresponding spatial locations across the 2D and 3D attention maps, contrary to the text's claim that it 'promotes consistency among the intersecting 2D and 3D representations.' This means the stated mechanism for the JE loss is not supported by the equation as written, but it is not a case of the paper deriving its result from an equation that contains the result. The empirical gains could still come from the shared-parameter JAM and the entropy regularizer's tendency to sharpen each attention map; the baseline comparisons are not logically presupposed by the method. Therefore the circularity score is low.

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

The central claim rests mainly on the assumption that attention consistency between 2D and 3D can be learned from aligned FaceScape data and transferred to 2D-only inference. The method also relies on the chosen backbone pair (IRNet-18, PointNet++) and several heuristically set hyperparameters. No new physical or mathematical entities are introduced.

free parameters (6)
  • AdaFace margin m = 0.5
    Adjusted heuristically from recommended settings (Sec. V-C); this changes the decision boundary and is tuned on the dataset.
  • AdaFace additive margin h = 0.0
    Modulates the additive margin; set heuristically (Sec. V-C).
  • AdaFace adaptive margin concentration t_a = 0.01
    Controls adaptive margin concentration; set heuristically (Sec. V-C).
  • AdaFace embedding scale s = 64
    Calibrates the norm of the embedding vector; set heuristically (Sec. V-C).
  • Loss weights for L2D, L3D, LJE = equal (1,1,1)
    Determined empirically with no sensitivity analysis (Sec. IV-E).
  • Entropy binning parameters = unspecified
    Attention maps are discretized into bins but the number and range of bins are never specified, so L_JE is not fully defined (Sec. IV-E).
assumptions (5)
  • domain assumption 3D point clouds are pose invariant after frontal alignment and sampling.
    The paper manually aligns meshes to frontal and rotates them before sampling (Sec. V-C), making the 3D branch independent of 2D pose by construction rather than by learned invariance.
  • domain assumption PointNet++ on 1024 downsampled points preserves identity-discriminative shape information.
    The method assumes the 3D representation is sufficient for face recognition (Sec. V-C, Table II).
  • ad hoc to paper Joint entropy minimization over attention maps improves cross-modal consistency.
    This is the paper's central heuristic (Sec. IV-E); as written it is mathematically false because the joint distribution factorizes, so the loss cannot enforce consistency.
  • domain assumption Shape-from-shading cues in 2D images are sufficient to infer 3D geometry for pose invariant FR.
    The motivation relies on monocular 3D inference being feasible (Sec. III, Eq. 1).
  • domain assumption FaceScape 2D-3D correspondences are accurate after reprojection.
    Aligned 2D images and 3D point clouds are paired by ID, expression, and pose; the framework depends on this correspondence being correct (Sec. V-C).

how reviews work

0 comments
Cite this review

Pith. "Pith review of 2D-3D Attention and Entropy for Pose Robust 2D Facial Recognition." pith.science (2026). https://pith.science/paper/H22Q3BZ2

@misc{pith2026250509073,
  author       = {Pith},
  title        = {Pith review of: 2D-3D Attention and Entropy for Pose Robust 2D Facial Recognition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H22Q3BZ2}},
  note         = {Machine review of arXiv:2505.09073}
}
abstract

Despite recent advances in facial recognition, there remains a fundamental issue concerning degradations in performance due to substantial perspective (pose) differences between enrollment and query (probe) imagery. Therefore, we propose a novel domain adaptive framework to facilitate improved performances across large discrepancies in pose by enabling image-based (2D) representations to infer properties of inherently pose invariant point cloud (3D) representations. Specifically, our proposed framework achieves better pose invariance by using (1) a shared (joint) attention mapping to emphasize common patterns that are most correlated between 2D facial images and 3D facial data and (2) a joint entropy regularizing loss to promote better consistency$\unicode{x2014}$enhancing correlations among the intersecting 2D and 3D representations$\unicode{x2014}$by leveraging both attention maps. This framework is evaluated on FaceScape and ARL-VTF datasets, where it outperforms competitive methods by achieving profile (90$\unicode{x00b0}$$\unicode{x002b}$) TAR @ 1$\unicode{x0025}$ FAR improvements of at least 7.1$\unicode{x0025}$ and 1.57$\unicode{x0025}$, respectively.

Figures

Figures reproduced from arXiv: 2505.09073 by the authors.

Figure 1
Figure 1. Conventional 2D image embedding networks for [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Our FR framework models pose-invariance by introducing JAM and JE, where this combination promotes consistency [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. ROC Curves for frontal and profile poses: (a) Frontal [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Comparison of attention maps superimposed on facial [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

81 extracted references · 69 canonical work pages

  1. [1]

    Afham, I

    M. Afham, I. Dissanayake, D. Dissanayake, A. Dharmasiri, K. Thi- lakarathna, and R. Rodrigo. Crosspoint: Self-supervised cross-modal contrastive learning for 3d point cloud understanding. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9902–9912, 2022

    2022

  2. [2]

    Alansari, O

    M. Alansari, O. A. Hay, S. Javed, A. Shoufan, Y . Zweiri, and N. Werghi. Ghostfacenets: Lightweight face recognition model from cheap operations. IEEE Access, 11:35429–35446, 2023

    work page 2023

  3. [3]

    Arsomngern, S

    P. Arsomngern, S. Nutanong, and S. Suwajanakorn. Learning geometric-aware properties in 2d representation using lightweight cad models, or zero real 3d pairs. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition , pages 21371– 21381, 2023

    work page 2023

  4. [4]

    Robust 3D face recognition in presence of pose and partial occlusions or missing parts

    P. Bagchi, D. Bhattacharjee, and M. Nasipuri. Robust 3d face recognition in presence of pose and partial occlusions or missing parts. arXiv preprint arXiv:1408.3709 , 2014

    work page Pith review arXiv 2014

  5. [5]

    Basri, D

    R. Basri, D. Jacobs, and I. Kemelmacher. Photometric stereo with general, unknown lighting. International Journal of computer vision , 72:239–257, 2007

    work page 2007

  6. [6]

    A. R. Bhople, A. M. Shrivastava, and S. Prakash. Point cloud based deep convolutional neural network for 3d face recognition. Multimedia Tools and Applications, 80(20):30237–30259, 2021

    work page 2021

  7. [7]

    Booth, E

    J. Booth, E. Antonakos, S. Ploumpis, G. Trigeorgis, Y . Panagakis, and S. Zafeiriou. 3d face morphable models ”in-the-wild”. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , pages 5464–5473, 2017

    work page 2017

  8. [8]

    Boutros, J

    F. Boutros, J. H. Grebe, A. Kuijper, and N. Damer. Idiff-face: Synthetic-based face recognition through fizzy identity-conditioned diffusion model. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 19650–19661, 2023

    work page 2023

Show all 81 references
  1. [9]

    K. Cao, Y . Rong, C. Li, X. Tang, and C. C. Loy. Pose-robust face recognition via deep residual equivariant mapping. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 5187–5196, 2018

    2018

  2. [10]

    N. Chen, L. Chu, H. Pan, Y . Lu, and W. Wang. Self-supervised image representation learning with geometric set consistency. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 19292–19302, 2022

    2022

  3. [11]

    Creswell, T

    A. Creswell, T. White, V . Dumoulin, K. Arulkumaran, B. Sengupta, and A. A. Bharath. Generative adversarial networks: An overview. IEEE signal processing magazine , 35(1):53–65, 2018

    2018

  4. [12]

    Dai and M

    A. Dai and M. Nießner. 3dmv: Joint 3d-multi-view prediction for 3d semantic scene segmentation. In Proceedings of the European Conference on Computer Vision (ECCV) , pages 452–468, 2018

    2018

  5. [13]

    J. Deng, J. Guo, N. Xue, and S. Zafeiriou. Arcface: Additive angular margin loss for deep face recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 4690– 4699, 2019

    2019

  6. [14]

    Deng and L

    Z. Deng and L. Jan Latecki. Amodal detection of 3d objects: Inferring 3d bounding boxes from 2d ones in rgb-depth images. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 5762–5770, 2017

    2017

  7. [15]

    X. Di, B. S. Riggan, S. Hu, N. J. Short, and V . M. Patel. Multi-scale thermal to visible face verification via attribute guided synthesis. IEEE Transactions on Biometrics, Behavior, and Identity Science, 3(2):266– 280, 2021

    2021

  8. [16]

    Eigen, C

    D. Eigen, C. Puhrsch, and R. Fergus. Depth map prediction from a single image using a multi-scale deep network. Advances in neural information processing systems , 27, 2014

    2014

  9. [17]

    C. N. Fondje, S. Hu, and B. S. Riggan. Learning domain and pose invariance for thermal-to-visible face recognition. IEEE Transactions on Biometrics, Behavior, and Identity Science , 5(1):15–28, 2022

    2022

  10. [18]

    C. N. Fondje, S. Hu, N. J. Short, and B. S. Riggan. Cross-domain identification for thermal-to-visible face recognition. In 2020 IEEE International Joint Conference on Biometrics (IJCB), pages 1–9. IEEE, 2020

    2020

  11. [19]

    Genova, X

    K. Genova, X. Yin, A. Kundu, C. Pantofaru, F. Cole, A. Sud, B. Brewington, B. Shucker, and T. Funkhouser. Learning 3d semantic segmentation with only 2d image supervision. In 2021 International Conference on 3D Vision (3DV) , pages 361–372. IEEE, 2021

    2021

  12. [20]

    Godard, O

    C. Godard, O. Mac Aodha, and G. J. Brostow. Unsupervised monocular depth estimation with left-right consistency. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 270–279, 2017

    2017

  13. [21]

    Gretton, K

    A. Gretton, K. M. Borgwardt, M. J. Rasch, B. Sch ¨olkopf, and A. Smola. A kernel two-sample test. The Journal of Machine Learning Research, 13(1):723–773, 2012

    2012

  14. [22]

    Hassner, S

    T. Hassner, S. Harel, E. Paz, and R. Enbar. Effective face frontalization in unconstrained images. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 4295–4304, 2015

    2015

  15. [23]

    K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 770–778, 2016

    2016

  16. [24]

    Hinton, O

    G. Hinton, O. Vinyals, and J. Dean. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 , 2015

    2015 arXiv

  17. [25]

    J. Ho, A. Jain, and P. Abbeel. Denoising diffusion probabilistic models. Advances in neural information processing systems , 33:6840–6851, 2020

    2020

  18. [26]

    J. Hou, X. Dai, Z. He, A. Dai, and M. Nießner. Mask3d: Pre- training 2d vision transformers by learning masked 3d priors. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13510–13519, 2023

    2023

  19. [27]

    J. Hou, S. Xie, B. Graham, A. Dai, and M. Nießner. Pri3d: Can 3d priors help 2d representation learning? In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 5693– 5702, 2021

    2021

  20. [28]

    W. Hu, H. Zhao, L. Jiang, J. Jia, and T.-T. Wong. Bidirectional projection network for cross dimension scene understanding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14373–14382, 2021

    2021

  21. [29]

    Jabberi, A

    M. Jabberi, A. Wali, B. Neji, T. Beyrouthy, and A. M. Alimi. Face shapenets for 3d face recognition. IEEE Access, 2023

    2023

  22. [30]

    Jaritz, T.-H

    M. Jaritz, T.-H. Vu, R. d. Charette, E. Wirbel, and P. P ´erez. xmuda: Cross-modal unsupervised domain adaptation for 3d semantic segmen- tation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 12605–12614, 2020

    2020

  23. [31]

    L. Jing, L. Zhang, and Y . Tian. Self-supervised feature learning by cross-modality and cross-view correspondences. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1581–1591, 2021

    2021

  24. [32]

    I. A. Kakadiaris, G. Passalis, G. Toderici, M. N. Murtuza, Y . Lu, N. Karampatziakis, and T. Theoharis. Three-dimensional face recog- nition in the presence of facial expressions: An annotated deformable model approach. IEEE Transactions on pattern analysis and machine intelli...

    2007

  25. [33]

    Kansy, A

    M. Kansy, A. Ra ¨el, G. Mignone, J. Naruniec, C. Schroers, M. Gross, and R. M. Weber. Controllable inversion of black-box face recognition models via diffusion. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 3167–3177, 2023

    2023

  26. [34]

    D. Kim, M. Hernandez, J. Choi, and G. Medioni. Deep 3d face iden- tification. In 2017 IEEE international joint conference on biometrics (IJCB), pages 133–142. IEEE, 2017

    2017

  27. [35]

    M. Kim, A. K. Jain, and X. Liu. Adaface: Quality adaptive margin for face recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 18750–18759, 2022

    2022

  28. [36]

    Kundu, Y

    A. Kundu, Y . Li, and J. M. Rehg. 3d-rcnn: Instance-level 3d object reconstruction via render-and-compare. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 3559– 3568, 2018

    2018

  29. [37]

    Kweon and K.-J

    H. Kweon and K.-J. Yoon. Joint learning of 2d-3d weakly supervised semantic segmentation. Advances in Neural Information Processing Systems, 35:30499–30511, 2022

    2022

  30. [38]

    Y . Lei, M. Bennamoun, M. Hayat, and Y . Guo. An efficient 3d face recognition approach using local geometrical signatures. Pattern Recognition, 47(2):509–524, 2014

    2014

  31. [39]

    M. Li, B. Huang, and G. Tian. A comprehensive survey on 3d face recognition methods. Engineering Applications of Artificial Intelligence, 110:104669, 2022

    2022

  32. [40]

    attention

    X. Li, Z. Li, H. Yang, G. Zhao, and L. Yin. Your “attention” deserves attention: A self-diversified multi-channel attention for facial action analysis. In 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2021) , pages 01–08. IEEE, 2021

    2021

  33. [41]

    Y . Li, R. Bu, M. Sun, W. Wu, X. Di, and B. Chen. Pointcnn: Convolution on x-transformed points. Advances in neural information processing systems, 31, 2018

    2018

  34. [42]

    Z. Li, Z. Chen, A. Li, L. Fang, Q. Jiang, X. Liu, J. Jiang, B. Zhou, and H. Zhao. Simipu: Simple 2d image and 3d point cloud unsupervised pre-training for spatial-aware visual representations. In Proceedings of the AAAI Conference on Artificial Intelligence , volume 36, pages ...

    2022

  35. [43]

    W. Liu, Y . Wen, Z. Yu, M. Li, B. Raj, and L. Song. Sphereface: Deep hypersphere embedding for face recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 212–220, 2017

    2017

  36. [44]

    Liu, Y .-K

    Y .-C. Liu, Y .-K. Huang, H.-Y . Chiang, H.-T. Su, Z.-Y . Liu, C.-T. Chen, C.-Y . Tseng, and W. H. Hsu. Learning from 2d: Contrastive pixel-to-point knowledge transfer for 3d pretraining. arXiv preprint arXiv:2104.04687, 2021

    2021 arXiv

  37. [45]

    Z. Liu, X. Qi, and C.-W. Fu. 3d-to-2d distillation for indoor scene parsing. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 4464–4474, 2021

    2021

  38. [46]

    M. Long, Y . Cao, J. Wang, and M. Jordan. Learning transferable features with deep adaptation networks. In International conference on machine learning , pages 97–105. PMLR, 2015

    2015

  39. [47]

    I. Masi, A. T. Tr ˜an, T. Hassner, J. T. Leksut, and G. Medioni. Do we really need to collect millions of faces for effective face recognition? In Computer Vision–ECCV 2016: 14th European Conference, Amster- dam, The Netherlands, October 11-14, 2016, Proceedings, Part V 14 , p...

    2016

  40. [48]

    Mpiperis, S

    I. Mpiperis, S. Malassiotis, and M. G. Strintzis. 3-d face recogni- tion with the geodesic polar representation. IEEE Transactions on Information Forensics and Security , 2(3):537–547, 2007

    2007

  41. [49]

    N. G. Nair, K. Mei, and V . M. Patel. At-ddpm: Restoring faces degraded by atmospheric turbulence using denoising diffusion proba- bilistic models. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , pages 3434–3443, 2023

    2023

  42. [50]

    Y . Nie, A. Dai, X. Han, and M. Nießner. Learning 3d scene priors with 2d supervision. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 792–802, 2023

    2023

  43. [51]

    Pecoraro, V

    R. Pecoraro, V . Basile, and V . Bono. Local multi-head channel self- attention for facial expression recognition. Information, 13(9):419, 2022

    2022

  44. [52]

    A. V . Phan, M. Le Nguyen, Y . L. H. Nguyen, and L. T. Bui. Dgcnn: A convolutional neural network over large-scale labeled graphs. Neural Networks, 108:533–543, 2018

    2018

  45. [53]

    Poster, M

    D. Poster, M. Thielke, R. Nguyen, S. Rajaraman, X. Di, C. N. Fondje, V . M. Patel, N. J. Short, B. S. Riggan, N. M. Nasrabadi, et al. A large-scale, time-synchronized visible and thermal face dataset. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer...

    2021

  46. [54]

    Qawaqneh, A

    Z. Qawaqneh, A. A. Mallouh, and B. D. Barkana. Deep convolutional neural network for age estimation based on vgg-face model. arXiv preprint arXiv:1709.01664, 2017

    2017 arXiv

  47. [55]

    C. R. Qi, L. Yi, H. Su, and L. J. Guibas. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in neural information processing systems , 30, 2017

    2017

  48. [56]

    H. Qin, R. Gong, X. Liu, M. Shen, Z. Wei, F. Yu, and J. Song. Forward and backward information retention for accurate binary neural networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 2250–2259, 2020

    2020

  49. [57]

    Schroff, D

    F. Schroff, D. Kalenichenko, and J. Philbin. Facenet: A unified embedding for face recognition and clustering. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 815–823, 2015

    2015

  50. [58]

    Sengupta, J.-C

    S. Sengupta, J.-C. Chen, C. Castillo, V . M. Patel, R. Chellappa, and D. W. Jacobs. Frontal to profile face verification in the wild. In 2016 IEEE winter conference on applications of computer vision (WACV) , pages 1–9. IEEE, 2016

    2016

  51. [59]

    Simonyan and A

    K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 , 2014

    2014 arXiv

  52. [60]

    Sun and K

    B. Sun and K. Saenko. Deep coral: Correlation alignment for deep domain adaptation. In Computer Vision–ECCV 2016 Workshops: Ams- terdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part III 14, pages 443–450. Springer, 2016

    2016

  53. [61]

    Taigman, M

    Y . Taigman, M. Yang, M. Ranzato, and L. Wolf. Deepface: Closing the gap to human-level performance in face verification. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 1701–1708, 2014

    2014

  54. [62]

    L. Tran, X. Yin, and X. Liu. Disentangled representation learning gan for pose-invariant face recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 1415– 1424, 2017

    2017

  55. [63]

    Tzeng, J

    E. Tzeng, J. Hoffman, K. Saenko, and T. Darrell. Adversarial discrim- inative domain adaptation. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 7167–7176, 2017

    2017

  56. [64]

    T.-H. Vu, H. Jain, M. Bucher, M. Cord, and P. P ´erez. Dada: Depth- aware domain adaptation in semantic segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 7364–7373, 2019

    2019

  57. [65]

    H. Wang, Y . Wang, Z. Zhou, X. Ji, D. Gong, J. Zhou, Z. Li, and W. Liu. Cosface: Large margin cosine loss for deep face recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5265–5274, 2018

    2018

  58. [66]

    Y . Wang, J. Liu, and X. Tang. Robust 3d face recognition by local shape difference boosting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(10):1858–1870, 2010

    2010

  59. [67]

    Y . Wen, K. Zhang, Z. Li, and Y . Qiao. A discriminative feature learning approach for deep face recognition. In B. Leibe, J. Matas, N. Sebe, and M. Welling, editors, Computer Vision – ECCV 2016 , pages 499–515, Cham, 2016. Springer International Publishing

    2016

  60. [68]

    J. Wu, C. Zhang, X. Zhang, Z. Zhang, W. T. Freeman, and J. B. Tenenbaum. Learning shape priors for single-view 3d completion and reconstruction. In Proceedings of the European conference on computer vision (ECCV) , pages 646–662, 2018

    2018

  61. [69]

    Y . Wu, M. Xing, Y . Zhang, Y . Xie, J. Fan, Z. Shi, and Y . Qu. Cross- modal unsupervised domain adaptation for 3d semantic segmentation via bidirectional fusion-then-distillation. In Proceedings of the 31st ACM International Conference on Multimedia , pages 490–498, 2023

    2023

  62. [70]

    C. Xu, S. Yang, T. Galanti, B. Wu, X. Yue, B. Zhai, W. Zhan, P. Vajda, K. Keutzer, and M. Tomizuka. Image2point: 3d point- cloud understanding with 2d image pretrained models. In European Conference on Computer Vision , pages 638–656. Springer, 2022

    2022

  63. [71]

    X. Yan, J. Gao, C. Zheng, C. Zheng, R. Zhang, S. Cui, and Z. Li. 2dpass: 2d priors assisted semantic segmentation on lidar point clouds. In European Conference on Computer Vision, pages 677–695. Springer, 2022

    2022

  64. [72]

    X. Yan, J. Yang, E. Yumer, Y . Guo, and H. Lee. Perspective trans- former nets: Learning single-view 3d object reconstruction without 3d supervision. Advances in neural information processing systems , 29, 2016

    2016

  65. [73]

    X. Yan, H. Zhan, C. Zheng, J. Gao, R. Zhang, S. Cui, and Z. Li. Let images give you more: Point cloud cross-modal training for shape analysis. Advances in Neural Information Processing Systems , 35:32398–32411, 2022

    2022

  66. [74]

    H. Yang, H. Zhu, Y . Wang, M. Huang, Q. Shen, R. Yang, and X. Cao. Facescape: a large-scale high quality 3d face dataset and detailed riggable 3d face prediction. In Proceedings of the ieee/cvf conference on computer vision and pattern recognition , pages 601–610, 2020

    2020

  67. [75]

    Yin and X

    X. Yin and X. Liu. Multi-task convolutional neural network for pose- invariant face recognition. IEEE Transactions on Image Processing , 27(2):964–975, 2017

    2017

  68. [76]

    M. Yuan, K. Fu, Z. Li, Y . Meng, and M. Wang. Pointmbf: A multi- scale bidirectional fusion network for unsupervised rgb-d point cloud registration. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 17694–17705, 2023

    2023

  69. [77]

    A. R. Zamir, A. Sax, W. Shen, L. J. Guibas, J. Malik, and S. Savarese. Taskonomy: Disentangling task transfer learning. InProceedings of the IEEE conference on computer vision and pattern recognition , pages 3712–3722, 2018

    2018

  70. [78]

    Zhang, Z

    B. Zhang, Z. Wang, Y . Ling, Y . Guan, S. Zhang, and W. Li. Mx2m: masked cross-modality modeling in domain adaptation for 3d semantic segmentation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 3401–3409, 2023

    2023

  71. [79]

    Zhang, V

    H. Zhang, V . M. Patel, B. S. Riggan, and S. Hu. Generative adversarial network-based synthesis of visible faces from polarimetrie thermal faces. In 2017 IEEE International Joint Conference on Biometrics (IJCB), pages 100–107. IEEE, 2017

    2017

  72. [80]

    J. Zhao, Y . Cheng, Y . Xu, L. Xiong, J. Li, F. Zhao, K. Jayashree, S. Pranata, S. Shen, J. Xing, et al. Towards pose invariant face recognition in the wild. In Proceedings of the IEEE conference on computer vision and pattern recognition , pages 2207–2216, 2018

    2018

  73. [81]

    Zhuang, L

    W. Zhuang, L. Chen, C. Hong, Y . Liang, and K. Wu. Ft-gan: face transformation with key points alignment for pose-invariant face recognition. Electronics, 8(7):807, 2019

    2019

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.