Hard-Mining Loss based Convolutional Neural Network for Face Recognition

Shiv Ram Dubey; Vaishnav Murali; Yash Srivastava

REVIEW 3 major objections 4 minor 35 references

Hard-Mining Loss based Convolutional Neural Network for Face Recognition

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Multiplying any face-recognition loss by a sigmoid of its own value shifts training weight to hard examples and improves verification accuracy.

desk verdict A minor focal-loss variant with an ambiguous formal definition and thin experiments; worth a conditional referee if the authors can fix the equations and add the missing baseline. read the letter →

arxiv 1908.09747 v2 pith:375IH2QU submitted 2019-08-09 cs.CV cs.LG

Yash Srivastava , Vaishnav Murali , Shiv Ram Dubey This is my paper · ORCID

classification cs.CVcs.LG

keywords facerecognitionhard-mininglossfunctiondeeplearningsigmoidweightingcross-entropyangularsoftmaxarcface

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

Face-recognition losses are usually dominated by easy examples, which are plentiful, so hard examples contribute little to the gradient. This paper proposes a generic wrapper that multiplies any loss $L$ by $\sigma(\beta L)$, a sigmoidal factor that grows with the loss value, so hard examples get up-weighted and easy ones are down-weighted. The paper argues this wrapper is loss-agnostic and demonstrates it by wrapping Cross-Entropy, Angular-Softmax, and ArcFace losses, training ResNet18 on CASIA-WebFace or MS-Celeb-1M, and testing on LFW and YTF. Across the reported settings, the wrapped losses match or exceed their unwrapped baselines in verification accuracy. The underlying claim is that simply reweighting by loss magnitude is enough to make a network learn hard-example characteristics.

What carries the argument

The load-bearing object is the Hard-Mining transform $L \mapsto \alpha L \sigma(\beta L)$, where $\sigma$ is a shifted logistic sigmoid with slope and center $A$, $B$. It carries the argument by turning the base loss value itself into a per-sample importance weight, so no explicit hard-sample search or pair selection is needed. Because the factor is monotone in $L$, samples that the current model finds difficult receive larger gradients automatically, and the definition of 'hard' shifts across training as the model improves.

What would settle it

Re-run the CASIA-WebFace to LFW comparison with the same code but set $B$ to 0.5 and again to 1.0, leaving every other setting fixed. If the Hard-Mining losses do not beat their unwrapped baselines under either neighboring $B$ value, the reported boost is tied to the exact hand-set hyperparameters rather than to the reweighting mechanism.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that loss-value-dependent reweighting is an effective and portable mechanism for hard-example emphasis in face recognition. The proposed Hard-Mining loss is defined as $L_{HM} = \alpha L \sigma(\beta L)$, with $\sigma(x) = (1+e^{-A(x-B)})^{-1}$, so an example with a large base loss is multiplied by a factor near one while an example with a small loss is multiplied by a factor near zero. With $\alpha=1.5$, $\beta=1.1$, $A=35$, and $B=0.75$, the wrapper lifts LFW accuracy from 95.35% to 96.75% for Cross-Entropy, from 97.12% to 97.3% for Angular-Softmax, and from 97.79% to 97.9% for ArcFace when trained on CASIA-WebFace; similar or smaller gains are reported on YTF and on MS-Celeb-1M training.

Load-bearing premise

The load-bearing premise is that the single hand-chosen set of hyperparameters $\alpha=1.5$, $\beta=1.1$, $A=35$, $B=0.75$ works across all three base losses and both training datasets without per-loss tuning.

Editorial extensions

If this is right

The wrapper is loss-agnostic, so the same Hard-Mining formula can be applied to margin-based losses beyond the three tested ones.
The reported accuracy gains, while small, appear on both test datasets and both training sets, so the effect is not tied to one benchmark.
Since the weighting uses only the current loss value, the method needs no separate hard-sample mining stage during training.
In most reported cells the Hard-Mining version is better than or equal to its base loss, supporting the paper's conclusion that easy examples dominate without reweighting.

Reading between the lines

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

An implication the paper leaves implicit: because the wrapper uses only the scalar loss value, the same formula should apply to other classification tasks where easy samples dominate, such as object detection; this is not demonstrated here.
The paper fixes four hyperparameters without an ablation, so a natural extension is to map how the gain varies with the sigmoid center $B$; that map could reveal whether the mechanism or the tuning produces the reported boost.
The sigmoid reweighting is one instance of the broader idea of confidence-based loss scaling, so the reported gains are likely to generalize to other confidence-based losses.

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.

Desk Editor's Note

read the letter

The short version: this paper reweights a base loss by a sigmoid of the loss itself, which is a close relative of focal loss, but it never cites or compares against focal loss. Worse, the formal definition has a batch-average vs per-sample ambiguity that could mean the method does not actually do what it claims. The experiments show small improvements that are sometimes negative, so the central "boosting" claim is overstated. Still, the idea is simple and the numbers are reported transparently.

What's genuinely new: the specific choice of a sigmoid on the loss value as a hard-mining multiplier. Focal loss down-weights easy examples via a probability-based term; this uses a sigmoid on the scalar loss. That is a cosmetic difference if the loss is per-sample, and a no-op if it is batch-averaged. The paper does not clarify which.

What it does well: the writing is clear, the benchmarks are standard, and the results are not cherry-picked—Table 2 shows two negative results on YTF that the conclusion glosses over but the data still contains. Using three base losses is a reasonable test bed.

Soft spots:

1. The formal ambiguity is load-bearing. Equations (2) and (4) define L as the batch-averaged loss in (1). If that is the intended reading, sigma(beta*L) is a constant for the whole batch, all per-sample gradients are scaled identically, and no hard mining occurs. Algorithm 1 also multiplies by beta, giving alpha*beta*L*sigma(...) rather than alpha*L*sigma(...). If per-sample losses are intended, the equations need to say so explicitly. As written, the formal definition does not implement the claimed mechanism—the stress-test note is correct.

2. No comparison to focal loss, the obvious prior. The cited hard-mining works [2,21,32] are related but none down-weight easy examples through a continuous multiplier on the loss itself. Focal loss is the closest and its omission is a real gap.

3. Empirical support is thin: no error bars, no repeated runs, and the four hyperparameters are picked without ablations or a validation set. The gains on LFW/CASIA are up to 1.4 points for Cross-Entropy, but for Angular-Softmax and ArcFace they are 0.1–0.2 points, and the MS-Celeb-to-YTF results are negative in two of three cases.

Who this is for: a reader working on loss design for face recognition might skim it as a related method. It deserves a serious referee because the formal issue is exactly the kind of thing a review should catch, and the missing focal-loss baseline is fixable. My recommendation: conditional accept, requiring the authors to clarify the per-sample definition, add a focal-loss comparison, and include a sensitivity analysis for alpha, beta, A, and B. Without that clarity, the contribution is hard to distinguish from global loss scaling.

Referee Report

3 major / 4 minor

Summary. The paper proposes a family of loss functions, called Hard-Mining loss, defined by multiplying an existing loss L by a sigmoid of L: L_HM = α L σ(βL), with σ(x)=1/(1+exp(-A(x-B))). The mechanism is intended to up-weight hard examples and down-weight easy examples during training. The authors apply the wrapper to Cross-Entropy, Angular-Softmax, and ArcFace losses, training ResNet18 on CASIA-WebFace or MS-Celeb-1M and evaluating verification accuracy on LFW and YTF. The central claim, stated in the abstract and conclusion, is that existing loss functions are boosted when used inside the proposed framework. The paper reports improvements in most of the six base-loss/dataset comparisons from Table 1 and Table 2, while acknowledging that some results are comparable.

Significance. If realized correctly, the proposed loss is simple and generic: it is a per-sample weighting scheme that requires no pair selection or external mining procedure and is applicable to any differentiable base loss. That feature is a genuine strength, and Figure 1 gives an intuitive illustration of the intended weighting behavior. However, the contribution is currently not established: the formal equations do not unambiguously define a per-sample reweighting, the algorithm contradicts the equations, the hyperparameters are chosen without sensitivity analysis, and the experimental support consists of single-run accuracy differences that are small and, in three cases, negative. The paper is therefore a promising idea with unsupported claims, and the central mechanism needs to be clarified and re-validated before the proposed loss can be accepted as a reliable contribution.

major comments (3)

[Section 2.2, Eqs. (2) and (4), and Algorithm 1] The formal definition does not implement the claimed per-example mechanism. Equation (1) defines L_CE as a batch-averaged scalar, and Eq. (4) substitutes that scalar into σ(βL_CE). Under that reading, the reweighting factor is identical for every sample in the batch, so the gradient of L_HM_CE is a common multiple of the gradient of L_CE, and the loss cannot "increase the loss for harder examples and decrease it for easier examples." If per-sample losses l_i are intended, the equations must introduce them explicitly, e.g., L_HM_CE = (1/N) Σ_i α l_i σ(β l_i), and the notation must be changed consistently. In addition, Algorithm 1 line 5 returns α × x × z with x = β × L_CE, which equals α β L_CE σ(...), not α L_CE σ(...) as in Eq. (4); the extra factor β makes Algorithm 1 inconsistent with the equation. This ambiguity is load-bearing because the central claim depends on the sigmoid factor varying across samples.
[Section 3.4 and Fig. 1] The four hyperparameters α=1.5, β=1.1, A=35, and B=0.75 are set empirically with no ablation and no validation split. With A=35 and B=0.75, σ is effectively a near-step function at L≈0.75, so the proposed loss is essentially a hard threshold imposed on the per-sample loss; small changes in B or the loss scale could drastically change which samples are mined. Because the same four values are assumed to work across all three base losses and both training datasets without any sensitivity analysis, the comparisons in Tables 1 and 2 reflect one hand-tuned configuration rather than a validated, generic algorithm. An ablation study or a sensitivity table varying α, β, A, and B is necessary to support the claim that the method is generic.
[Section 4, Tables 1 and 2] The abstract and conclusion claim that existing losses are "boosted" by the proposed framework, but the data in Table 2 contradict that in three of six comparisons: LHM_AS on YTF drops from 94.1 to 93.8, LHM_AF on YTF drops from 95.1 to 94.9, and LHM_CE on LFW ties at 95.1. The positive differences elsewhere are at most 0.4 percentage points (e.g., LHM_AF on LFW, 98.0 vs 97.6), and the paper reports no error bars, no repeated runs, and no statistical significance. Consequently, the evidence does not support the strong "boosted" claim; at best it shows comparable accuracy in a few configurations. The authors should either soften the claim or provide statistical validation and specify the exact evaluation protocol (e.g., standard 10-fold LFW/YTF verification protocol).

minor comments (4)

[Section 2.3] The text states that "σ is defined in (1)", but the sigmoid is defined in Eq. (3), not Eq. (1).
[Tables 1 and 2] There are typos in the table headers: "Angular-Sofmax" should be "Angular-Softmax" in Table 1 and "Cross-Entropy los" should be "Cross-Entropy loss" in Table 2.
[Throughout] Several minor grammatical and typographical errors should be corrected, including "intution" in Section 2.1, "perfomed" in Section 4, "we propose" capitalized mid-sentence in the abstract, and inconsistent use of "loss" vs "losses".
[Section 3.2] The description of the MS-Celeb-1M data is incomplete: the "cleaned and refined subset" should be specified precisely (e.g., a citation to the specific cleaned list or a description of the cleaning procedure), since the noise level of that dataset materially affects training outcomes.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the proposed loss is an explicit reweighting of base losses and the empirical claims rest on external benchmarks, with only minor non-load-bearing self-citation.

full rationale

The Hard-Mining loss is defined directly as L_HM = α × L × σ(β × L), so the stated hard/easy reweighting is part of the construction rather than a hidden reuse of the target accuracy. No parameter is fitted to the test accuracies in the paper: the hyperparameters α=1.5, β=1.1, A=35, B=0.75 are reported as empirically set, and the claimed boosts are measured on the external LFW and YTF benchmarks against independently evaluated Cross-Entropy, Angular-Softmax, and ArcFace baselines. The only self-citation is [23], used to say that ArcFace was found outstanding in the authors' prior comparison; that statement is contextual rather than load-bearing because ArcFace is also measured in Tables 1 and 2 of this paper. There is a real formal ambiguity: Eq. (1) defines L_CE as a batch average, and substituting that scalar into Eq. (4) would make the sigmoid multiplier identical for every sample, contradicting the claimed per-sample hard-mining mechanism, but this is a correctness or clarity concern, not a circular reduction of the results to their inputs.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The paper's central claim rests on four hyperparameters that are hand-tuned on the evaluation benchmarks, plus standard mathematical assumptions and a domain assumption about the validity of the LFW/YTF protocol. There are no new physical or structural entities introduced.

free parameters (4)

alpha = 1.5
Global multiplier for the hard-mining loss, set empirically in Section 3.4 without ablation.
beta = 1.1
Scales the base loss before applying the sigmoid; controls the sharpness of the reweighting. Set empirically.
A = 35
Sigmoid steepness parameter in Eq. 3; chosen to make the curve in Fig. 1 look the way the authors want.
B = 0.75
Sigmoid threshold that defines easy versus hard examples; with B=0.75 and A=35 the sigmoid is near zero for small losses, effectively discarding easy examples. Chosen empirically.

assumptions (3)

standard math Sigmoid and cross-entropy definitions are standard and correctly applied.
Equations (1)-(3) rely on standard definitions of softmax, cross-entropy, and the logistic sigmoid.
domain assumption LFW/YTF verification accuracy is a suitable performance measure for face recognition loss comparison.
The paper uses these benchmarks as ground truth for performance without questioning their validity for the claim.
domain assumption ResNet18 trained for 20 epochs with the stated schedule is a sufficient and fair protocol for comparing loss functions.
The paper assumes that this training setup gives a fair comparison among the base losses, but does not justify why more epochs or a different backbone would not change the conclusions.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Hard-Mining Loss based Convolutional Neural Network for Face Recognition." pith.science (2026). https://pith.science/paper/375IH2QU

@misc{pith2026190809747,
  author       = {Pith},
  title        = {Pith review of: Hard-Mining Loss based Convolutional Neural Network for Face Recognition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/375IH2QU}},
  note         = {Machine review of arXiv:1908.09747}
}

read the original abstract

Face Recognition is one of the prominent problems in the computer vision domain. Witnessing advances in deep learning, significant work has been observed in face recognition, which touched upon various parts of the recognition framework like Convolutional Neural Network (CNN), Layers, Loss functions, etc. Various loss functions such as Cross-Entropy, Angular-Softmax and ArcFace have been introduced to learn the weights of network for face recognition. However, these loss functions do not give high priority to the hard samples as compared to the easy samples. Moreover, their learning process is biased due to a number of easy examples compared to hard examples. In this paper, we address this issue by considering hard examples with more priority. In order to do so, We propose a Hard-Mining loss by increasing the loss for harder examples and decreasing the loss for easy examples. The proposed concept is generic and can be used with any existing loss function. We test the Hard-Mining loss with different losses such as Cross-Entropy, Angular-Softmax and ArcFace. The proposed Hard-Mining loss is tested over widely used Labeled Faces in the Wild (LFW) and YouTube Faces (YTF) datasets. The training is performed over CASIA-WebFace and MS-Celeb-1M datasets. We use the residual network (i.e., ResNet18) for the experimental analysis. The experimental results suggest that the performance of existing loss functions is boosted when used in the framework of the proposed Hard-Mining loss.

Figures

Figures reproduced from arXiv: 1908.09747 by the authors.

**Figure 1.** Loss value vs Likelihood (i.e., probability for correct class) plot for the Cross-Entropy loss and Hard-Mining loss functions. Note that the Hard-Mining loss is computed on the output of Cross-Entropy loss. no big difference between easy and hard examples. We believe that if the probability is more than 0.5 then the loss should be minimum. Whereas, if the probability is less than 0.5 then the loss should be on highe… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 27 canonical work pages

[1]

IEEE Signal Processing Letters 25(5), 635–639 (2018)

Chakraborti, T., McCane, B., Mills, S., Pal, U.: Loop descriptor: Local optimal-oriented pattern. IEEE Signal Processing Letters 25(5), 635–639 (2018)

work page 2018
[2]

In: 2018 25th IEEE International Conference on Image Processing (ICIP)

Chen, K., Chen, Y ., Han, C., Sang, N., Gao, C., Wang, R.: Improving person re-identiﬁcation by adaptive hard sample mining. In: 2018 25th IEEE International Conference on Image Processing (ICIP). pp. 1638–1642. IEEE (2018)

work page 2018
[3]

arXiv preprint arXiv:1801.07698 (2018)

Deng, J., Guo, J., Xue, N., Zafeiriou, S.: Arcface: Additive angular margin loss for deep face recognition. arXiv preprint arXiv:1801.07698 (2018)

arXiv 2018
[4]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops

Deng, J., Zhou, Y ., Zafeiriou, S.: Marginal loss for deep face recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. pp. 60–68 (2017) Hard-Mining Loss based Convolutional Neural Network for Face Recognition 9

work page 2017
[5]

Neural Com- puting and Applications pp

Dubey, S.R., Roy, S.K., Chakraborty, S., Mukherjee, S., Chaudhuri, B.B.: Local bit-plane decoded convolutional neural network features for biomedical image retrieval. Neural Com- puting and Applications pp. 1–13 (2019)

work page 2019
[6]

IEEE Transactions on Image Processing23(12), 5323–5333 (2014)

Dubey, S.R., Singh, S.K., Singh, R.K.: Rotation and illumination invariant interleaved inten- sity order-based local descriptor. IEEE Transactions on Image Processing23(12), 5323–5333 (2014)

work page 2014
[7]

IEEE Transactions on Image Processing 24(12), 5892–5903 (2015)

Dubey, S.R., Singh, S.K., Singh, R.K.: Local wavelet pattern: a new feature descriptor for image retrieval in medical ct databases. IEEE Transactions on Image Processing 24(12), 5892–5903 (2015)

work page 2015
[8]

IEEE transactions on image processing 25(9), 4018–4032 (2016)

Dubey, S.R., Singh, S.K., Singh, R.K.: Multichannel decoded local binary patterns for content-based image retrieval. IEEE transactions on image processing 25(9), 4018–4032 (2016)

work page 2016

Show all 35 references

[9]

In: European Conference on Computer Vision

Guo, Y ., Zhang, L., Hu, Y ., He, X., Gao, J.: Ms-celeb-1m: A dataset and benchmark for large- scale face recognition. In: European Conference on Computer Vision. pp. 87–102. Springer (2016)

2016
[10]

In: Pro- ceedings of the IEEE conference on computer vision and pattern recognition

He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Pro- ceedings of the IEEE conference on computer vision and pattern recognition. pp. 770–778 (2016)

2016
[11]

In: Proceedings of the IEEE conference on computer vision and pattern recognition

Hu, J., Shen, L., Sun, G.: Squeeze-and-excitation networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 7132–7141 (2018)

2018
[12]

Huang, G.B., Ramesh, M., Berg, T., Learned-Miller, E.: Labeled faces in the wild: A database for studying face recognition in unconstrained environments. Tech. Rep. 07-49, University of Massachusetts, Amherst (October 2007)

2007
[13]

IEEE Signal Processing Letters 26(1), 129–133 (2018)

Kou, Q., Cheng, D., Zhuang, H., Gao, R.: Cross-complementary local binary pattern for robust texture classiﬁcation. IEEE Signal Processing Letters 26(1), 129–133 (2018)

2018
[14]

In: Advances in neural information processing systems

Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classiﬁcation with deep convolutional neural networks. In: Advances in neural information processing systems. pp. 1097–1105 (2012)

2012
[15]

In: International Conference on Neural Information Processing

Liang, X., Wang, X., Lei, Z., Liao, S., Li, S.Z.: Soft-margin softmax for deep classiﬁca- tion. In: International Conference on Neural Information Processing. pp. 413–421. Springer (2017)

2017
[16]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recog- nition

Liu, H., Zhu, X., Lei, Z., Li, S.Z.: Adaptiveface: Adaptive margin and sampling for face recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recog- nition. pp. 11947–11956 (2019)

2019
[17]

In: Proceedings of the IEEE conference on computer vision and pattern recognition

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

2017
[18]

In: ICML

Liu, W., Wen, Y ., Yu, Z., Yang, M.: Large-margin softmax loss for convolutional neural networks. In: ICML. p. 7 (2016)

2016
[19]

Interna- tional journal of computer vision 115(3), 211–252 (2015)

Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al.: Imagenet large scale visual recognition challenge. Interna- tional journal of computer vision 115(3), 211–252 (2015)

2015
[20]

In: Proceedings of the IEEE conference on computer vision and pattern recognition

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

2015
[21]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops

Smirnov, E., Melnikov, A., Oleinik, A., Ivanova, E., Kalinovskiy, I., Luckyanets, E.: Hard example mining with auxiliary embeddings. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. pp. 37–46 (2018)

2018
[22]

IEEE Signal Processing Letters25(5), 625–629 (2018) 10 Yash Srivastava, Vaishnav Murali and Shiv Ram Dubey

Song, T., Xin, L., Gao, C., Zhang, G., Zhang, T.: Grayscale-inversion and rotation invariant texture description using sorted local gradient pattern. IEEE Signal Processing Letters25(5), 625–629 (2018) 10 Yash Srivastava, Vaishnav Murali and Shiv Ram Dubey

2018
[23]

arXiv preprint arXiv:1901.05903 (2019)

Srivastava, Y ., Murali, V ., Dubey, S.R.: A performance comparison of loss functions for deep face recognition. arXiv preprint arXiv:1901.05903 (2019)

2019 arXiv
[24]

In: Advances in neural information processing systems

Sun, Y ., Chen, Y ., Wang, X., Tang, X.: Deep learning face representation by joint identiﬁcation-veriﬁcation. In: Advances in neural information processing systems. pp. 1988– 1996 (2014)

2014
[25]

In: Proceedings of the IEEE conference on computer vision and pattern recognition

Szegedy, C., Liu, W., Jia, Y ., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V ., Rabinovich, A.: Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 1–9 (2015)

2015
[26]

In: Proceedings of the IEEE conference on computer vision and pattern recognition

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

2014
[27]

IEEE Signal Processing Letters 25(7), 926–930 (2018)

Wang, F., Cheng, J., Liu, W., Liu, H.: Additive margin softmax for face veriﬁcation. IEEE Signal Processing Letters 25(7), 926–930 (2018)

2018
[28]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

Wang, H., Wang, Y ., Zhou, Z., Ji, X., Gong, D., Zhou, J., Li, Z., Liu, W.: Cosface: Large margin cosine loss for deep face recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 5265–5274 (2018)

2018
[29]

IEEE Signal Processing Letters (2019)

Wang, Y ., Ward, R.K., Wang, Z.J.: Coarse-to-ﬁne image dehashing using deep pyramidal residual learning. IEEE Signal Processing Letters (2019)

2019
[30]

arXiv preprint arXiv:1805.06741 (2018)

Wei, X., Wang, H., Scotney, B., Wan, H.: Minimum margin loss for deep face recognition. arXiv preprint arXiv:1805.06741 (2018)

2018 arXiv
[31]

IEEE (2011)

Wolf, L., Hassner, T., Maoz, I.: Face recognition in unconstrained videos with matched back- ground similarity. IEEE (2011)

2011
[32]

arXiv preprint arXiv:1710.00478 (2017)

Xiao, Q., Luo, H., Zhang, C.: Margin sample mining loss: A deep learning based method for person re-identiﬁcation. arXiv preprint arXiv:1710.00478 (2017)

2017 arXiv
[33]

arXiv preprint arXiv:1411.7923 (2014)

Yi, D., Lei, Z., Liao, S., Li, S.Z.: Learning face representation from scratch. arXiv preprint arXiv:1411.7923 (2014)

2014 arXiv
[34]

IEEE Signal Processing Letters23(10), 1499–1503 (2016)

Zhang, K., Zhang, Z., Li, Z., Qiao, Y .: Joint face detection and alignment using multitask cascaded convolutional networks. IEEE Signal Processing Letters23(10), 1499–1503 (2016)

2016
[35]

In: 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018)

Zhou, Y ., Liu, D., Huang, T.: Survey of face detection on low-quality images. In: 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018). pp. 769–773. IEEE (2018)

2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

IEEE Signal Processing Letters 25(5), 635–639 (2018)

Chakraborti, T., McCane, B., Mills, S., Pal, U.: Loop descriptor: Local optimal-oriented pattern. IEEE Signal Processing Letters 25(5), 635–639 (2018)

work page 2018

[2] [2]

In: 2018 25th IEEE International Conference on Image Processing (ICIP)

Chen, K., Chen, Y ., Han, C., Sang, N., Gao, C., Wang, R.: Improving person re-identiﬁcation by adaptive hard sample mining. In: 2018 25th IEEE International Conference on Image Processing (ICIP). pp. 1638–1642. IEEE (2018)

work page 2018

[3] [3]

arXiv preprint arXiv:1801.07698 (2018)

Deng, J., Guo, J., Xue, N., Zafeiriou, S.: Arcface: Additive angular margin loss for deep face recognition. arXiv preprint arXiv:1801.07698 (2018)

arXiv 2018

[4] [4]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops

Deng, J., Zhou, Y ., Zafeiriou, S.: Marginal loss for deep face recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. pp. 60–68 (2017) Hard-Mining Loss based Convolutional Neural Network for Face Recognition 9

work page 2017

[5] [5]

Neural Com- puting and Applications pp

Dubey, S.R., Roy, S.K., Chakraborty, S., Mukherjee, S., Chaudhuri, B.B.: Local bit-plane decoded convolutional neural network features for biomedical image retrieval. Neural Com- puting and Applications pp. 1–13 (2019)

work page 2019

[6] [6]

IEEE Transactions on Image Processing23(12), 5323–5333 (2014)

Dubey, S.R., Singh, S.K., Singh, R.K.: Rotation and illumination invariant interleaved inten- sity order-based local descriptor. IEEE Transactions on Image Processing23(12), 5323–5333 (2014)

work page 2014

[7] [7]

IEEE Transactions on Image Processing 24(12), 5892–5903 (2015)

Dubey, S.R., Singh, S.K., Singh, R.K.: Local wavelet pattern: a new feature descriptor for image retrieval in medical ct databases. IEEE Transactions on Image Processing 24(12), 5892–5903 (2015)

work page 2015

[8] [8]

IEEE transactions on image processing 25(9), 4018–4032 (2016)

Dubey, S.R., Singh, S.K., Singh, R.K.: Multichannel decoded local binary patterns for content-based image retrieval. IEEE transactions on image processing 25(9), 4018–4032 (2016)

work page 2016