REVIEW 4 major objections 4 minor 28 references
Exploring the Feasibility of Deep Learning Techniques for Accurate Gender Classification from Eye Images
T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read A purpose-built convolutional neural network classifies gender from color periocular images at 99% accuracy on the CVBL dataset and 96% on a second dataset, with about 7.2 million parameters.
desk verdict A plausible but under-documented CNN result on periocular gender classification; the 99% CVBL claim hinges on an identity-disjoint split that the paper never describes. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the proposed CNN architecture: 3-by-3 convolutional filters with same padding and stride 1, batch normalization, exponential linear unit (ELU) activations, 2-by-2 max pooling, dropout at probability 0.15, one fully-connected output with a sigmoid, and binary cross-entropy loss trained with stochastic gradient descent for 200 epochs at batch size 128. With roughly 7.2 million total parameters, the architecture carries the argument by showing that high accuracy can be achieved without a large pretrained backbone and that the color periocular region supplies enough signal for the task.
What would settle it
Hold out the entire CVBL dataset until every architecture and hyperparameter decision is frozen, then run an identity-disjoint train/test split and measure accuracy on that untouched data; if the 99% figure depended on tuning or on same-person images appearing in both sets, the number would drop.
Extended reading notes
Core claim
The central claim is that the color periocular region alone carries enough gender-discriminative information for a compact custom CNN to reach near-ceiling accuracy: 99% on CVBL and 96% on Female and Male. The network is trained end-to-end with binary cross-entropy and a sigmoid output, using convolutional blocks with batch normalization, exponential linear unit activations, max pooling, and dropout. The authors interpret these results as showing that the periocular region is a reliable and efficient basis for gender classification, and they compare the model against earlier methods using iris texture, pretrained CNNs, and near-infrared images, reporting better accuracy in their test setup.
Load-bearing premise
The 99% CVBL result rests on the premise that the CVBL images were not used, directly or through model selection, to choose the architecture or hyperparameters, and that the train/test split keeps distinct identities apart.
Editorial extensions
If this is right
- If the reported accuracies generalize, periocular color images are sufficient for automated gender classification even when full facial images are unavailable.
- The compact parameter count implies that a task-specific CNN, rather than a large pretrained network, can serve as an efficient gender classifier in resource-constrained settings.
- The favorable comparison with iris-based and earlier periocular methods indicates that color periocular cues carry stronger gender information than iris texture alone.
- The CVBL dataset, described as previously unused, would become a natural benchmark for future periocular gender classification research.
Reading between the lines
- A direct extension the paper does not run is to vary cosmetics, disguise, and occlusion systematically on the same datasets, even though those factors motivate the study.
- The same architecture could be transferred to near-infrared periocular images to test whether the color-based advantage persists in the modality used by several comparison baselines.
- Breaking down accuracy by left and right eye, image resolution, and lighting or image-quality subgroups would reveal whether the headline numbers are uniform or driven by a subset of easy images.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes a custom CNN for gender classification from color periocular images and reports 99% accuracy on the CVBL dataset and 96% accuracy on the Kaggle (Female and Male) eye dataset, claiming superiority over prior methods. The version of the manuscript provided to me contains the abstract, introduction and related work, a background section on CNNs, a description of the proposed architecture with a parameter table, and references, but it does not include the dataset description, experimental setup, results section, or comparison tables that would support the headline accuracy figures.
Significance. If the reported accuracies are reproducible under a properly designed identity-disjoint evaluation, the paper would provide useful evidence that periocular color images support near-ceiling gender classification on these two datasets, and the explicit architecture and hyperparameter table in Table 2 would aid replication. However, the central claim currently rests on an under-specified evaluation protocol, and the manuscript as presented contains no results section that would allow the reader to verify the 99% and 96% numbers or the claimed comparison with state-of-the-art methods. The paper does not provide code, data, or error bars, so its main contribution is an empirical accuracy claim that cannot yet be independently assessed.
major comments (4)
- [Abstract and experimental protocol (missing)] The headline claim of 99% accuracy on the CVBL dataset cannot be evaluated because the manuscript does not state whether the train/test split is identity-disjoint, how many images were used for training and testing, whether classes are balanced, or what preprocessing and augmentation were applied. Reference [34] identifies CVBL as an iris gender classification database with multiple images per subject; if the split was done at the image level rather than the subject level, the same subject could appear in both partitions, and the network could memorize subjects rather than learn gender. The term 'previously unused' also leaves unclear whether CVBL was used for architecture or hyperparameter selection.
- [Results and comparison (absent from provided text)] The abstract and introduction state that the model was evaluated with an extensive range of metrics and compared with state-of-the-art approaches, but the provided manuscript contains no results section, no comparison table, no confusion matrix, no ROC or precision/recall values, and no confidence intervals. Without these elements, the claimed 99% and 96% accuracies and the stated superiority over prior methods are unsupported in the visible text. If this material exists in the full submission, it must be included in the reviewed version; as it stands, the central empirical claim is unverifiable.
- [Table 2] Table 2 reports the optimizer, loss, number of epochs, and batch size, but it omits the learning rate, momentum, weight decay, weight initialization scheme, image resolution, and number of repeated runs. With 200 epochs and no stated repetition strategy, the reported accuracy values have no variance estimate; a single favorable run could produce a 99% figure even if the average performance is lower. The manuscript should report mean and standard deviation over multiple runs with different seeds, or otherwise justify the stability of the headline numbers.
- [Introduction and related work] The introduction's claim that the paper provides 'a comprehensive comparison with other state-of-the-art approaches' is not backed by any visible evaluation against the cited prior works, such as the PCANET method [14] or the pretrained-CNN approach [13]. A proper comparison needs matched train/test protocols, identical preprocessing, and preferably the same backbone or at least reported parameter counts and training costs; none of these are presented in the visible text.
minor comments (4)
- [Table 2] Table 2 lists 'Batch Normalization Mini-batch size 64' while the Optimizer row lists 'Batch size 128'; please clarify whether the mini-batch size is 64 or 128 during training.
- [Abstract] The phrase 'small number of learnable parameters (7,235,089)' is misleading: 7.2 million parameters is not small in absolute terms, and the paper does not compare this count with the parameter counts of the prior models it discusses.
- [General presentation] The provided text jumps from the CNN background section directly to the proposed architecture and then to references, with no visible section for the datasets or experiments; if this is an extraction artifact, the submitted version should be checked for completeness, and if not, the manuscript is structurally incomplete.
- [References] Reference formatting is inconsistent, for example [13] and [25] contain stray brackets, and some entries use inconsistent journal-name capitalization; a final proofreading pass is needed.
Circularity Check
No circular derivation: the paper reports empirical CNN accuracies and contains no theory-to-prediction chain that reduces to its inputs.
full rationale
The paper is an empirical deep-learning study. Its central claims are measured classification accuracies (99% on CVBL, 96% on Female and Male) from a proposed CNN architecture. There are no equations from which results are derived, no fitted parameters renamed as predictions, no uniqueness theorem, and no self-citation chain carrying the load. The phrase 'previously unused CVBL dataset' is a claim about evaluation independence, but the visible text does not contain the dataset split, model-selection protocol, or class balance needed to establish or refute leakage; that is a methodology/reproducibility concern, not an instance of a result being equivalent to its input by definition. The acknowledged dependence of accuracy on dataset quality, resolution, and methodology is a limitation statement, not a circular step. Under the hard rule that circularity must be exhibited by quotation and specific reduction, no such reduction is present, so the appropriate finding is no significant circularity (score 0).
Assumptions & free parameters
free parameters (3)
- CNN architecture choices =
Table 2: 3x3 convolutions, ELU, max-pool 2x2, dropout 0.15, one sigmoid output
- Training hyperparameters =
200 epochs, batch size 128, SGD, binary cross-entropy; learning rate not stated in visible text
- Data preprocessing and split choices =
Not stated in the visible text
assumptions (3)
- domain assumption The periocular region contains enough gender information in color images for the reported accuracy.
- domain assumption Gender labels in CVBL and the Kaggle Female and Male dataset are accurate and class distributions are sufficiently balanced.
- domain assumption The CNN training procedure with SGD and the listed hyperparameters converges reliably for this task.
Cite this review
Pith. "Pith review of Exploring the Feasibility of Deep Learning Techniques for Accurate Gender Classification from Eye Images." pith.science (2026). https://pith.science/paper/VL7QXV3I
@misc{pith2026250800135,
author = {Pith},
title = {Pith review of: Exploring the Feasibility of Deep Learning Techniques for Accurate Gender Classification from Eye Images},
year = {2026},
howpublished = {\url{https://pith.science/paper/VL7QXV3I}},
note = {Machine review of arXiv:2508.00135}
}
read the original abstract
Gender classification has emerged as a crucial aspect in various fields, including security, human-machine interaction, surveillance, and advertising. Nonetheless, the accuracy of this classification can be influenced by factors such as cosmetics and disguise. Consequently, our study is dedicated to addressing this concern by concentrating on gender classification using color images of the periocular region. The periocular region refers to the area surrounding the eye, including the eyelids, eyebrows, and the region between them. It contains valuable visual cues that can be used to extract key features for gender classification. This paper introduces a sophisticated Convolutional Neural Network (CNN) model that utilizes color image databases to evaluate the effectiveness of the periocular region for gender classification. To validate the model's performance, we conducted tests on two eye datasets, namely CVBL and (Female and Male). The recommended architecture achieved an outstanding accuracy of 99% on the previously unused CVBL dataset while attaining a commendable accuracy of 96% with a small number of learnable parameters (7,235,089) on the (Female and Male) dataset. To ascertain the effectiveness of our proposed model for gender classification using the periocular region, we evaluated its performance through an extensive range of metrics and compared it with other state-of-the-art approaches. The results unequivocally demonstrate the efficacy of our model, thereby suggesting its potential for practical application in domains such as security and surveillance.
Reference graph
Works this paper leans on
-
[34]
Aryanmehr, S.; Karimi, M.; Boroujeni, F. Z. CVBL IRIS Gender Classification Database Image Processing and Biometric Research, Computer Vision and Biometric Laboratory (CVBL). 2018 3rd IEEE Int. Conf. Image, Vis. Comput. ICIVC 2018 2018, 433 –438. https://doi.org/10.1109/ICIVC.2018.8492757
-
[14]
Abdalrady, N. A.; Aly, S. Fusion of Multiple Simple Convolutional Neural Networks for Gender Classification. In 2020 International Conference on Innovative Trends in Communication and Computer Engineering (ITCE); 2020; pp 251 –256. https://doi.org/10.1109/ITCE48509.2020.9047798
arXiv 2020
-
[13]
CIMTAY, Y.; YILMAZ, G. N. Gender Classification from Eye Images by Using Pretrained Convolutional Neural Networks. Eurasia Proc. Sci. Technol. Eng. Math. 2021, 14 (March 2022), 39 –44. https://doi.org/10.55549/epstem.1050171
-
[1]
INTRODUCTION Gender prediction is a crucial aspect of human interaction as gender plays a significant role in social dynamics. It can be considered as a "soft biometric" feature in the field of biometrics, which refers to non- unique characteristics that can complement traditional biometrics such as fingerprints and iris scans [1]. The use of soft biometr...
work page Pith review arXiv 2025
-
[3]
CONVOLUTIONAL NEURAL NETWORKS The field of deep learning falls under the umbrella of machine learning, emphasizing the util ization of deep neural networks to derive high -level abstractions from data. In recent years, there has been a notable surge in applying deep learning methods to address diverse artificial intelligence challenges, including semantic...
-
[11]
La; Bisogni, C.; Cascone, L.; Narducci, F
Rocca, D. La; Bisogni, C.; Cascone, L.; Narducci, F. Periocular Data Fusion for Age and Gender Classification. J. Imaging 2022, Vol. 8, Page 307 2022, 8 (11), 307. https://doi.org/10.3390/JIMAGING8110307
-
[12]
Amri, R.; Gazdar, A.; Barhoumi, W. A Comparative Study on the Importance of Each Face Part in Facial Gender Recognition via Convolutional Neural Networks. In 2021 IEEE/ACS 18th International Co nference on Computer Systems and Applications (AICCSA) ; 2021; pp 1 –8. https://doi.org/10.1109/AICCSA53542.2021.9686825
-
[15]
Kumari, P.; Seeja, K. R. Periocular Biometrics for Non - Ideal Images: With off -the-Shelf Deep CNN & Transfer Learning Approach. Procedia Comput. Sci. 2020, 167, 344–352. https://doi.org/https://doi.org/10.1016/j.procs.2020.03.23 4
Show all 28 references
-
[16]
Relevant Features for Gender Classification in NIR Periocular Images
Viedma, I.; Tapia, J.; Iturriaga, A.; Busch, C. Relevant Features for Gender Classification in NIR Periocular Images. IET Biometrics 2019, 8 (5), 340 –350. https://doi.org/10.1049/IET-BMT.2018.5233
2019
-
[17]
Predicting Gender from Iris Texture May Be Harder than It Seems
Kuehlkamp, A.; Bowyer, K. Predicting Gender from Iris Texture May Be Harder than It Seems. Proc. - 2019 IEEE Winter Conf. Appl. Comput. Vision, WACV 2019 2019, 904–912. https://doi.org/10.1109/WACV.2019.00101
2019
-
[18]
Sex-Classification from Cellphones Periocular Iris Images
Tapia, J.; Arellano, C.; Viedma, I. Sex-Classification from Cellphones Periocular Iris Images. Adv. Comput. Vis. Pattern Recognit. 2019, 227 –242. https://doi.org/10.1007/978-3-030-26972-2_11
2019 doi
-
[19]
CNN -Based Gender Classification in near - Infrared Periocular Images
Manyala, A.; Cholakkal, H.; Anand, V.; Kanhangad, V.; Rajan, D. CNN -Based Gender Classification in near - Infrared Periocular Images. Pattern Anal. Appl. 2019, 22 (4), 1493 –1504. https://doi.org/10.1007/s10044 -018- 0722-3
2019 doi
-
[20]
Generalizable Deep Features for Ocular Biometrics
Reddy, N.; Rattani, A.; Derakh shani, R. Generalizable Deep Features for Ocular Biometrics. Image Vis. Comput. 2020, 103, 103996. https://doi.org/10.1016/J.IMAVIS.2020.103996
2020
-
[21]
Khalifa, N. E. M.; Taha, M. H. N.; Hassanien, A. E.; Mohamed, H. N. E. T. Deep Iris: Deep Learning for Gender Classification through Iris Patterns. Acta Inform. Medica 2019, 27 (2), 96 –102. https://doi.org/10.5455/aim.2019.27.96-102
2019 doi
-
[22]
C.; Roth, H
Shin, H. C.; Roth, H. R.; Gao, M.; Lu, L.; Xu, Z.; Nogues, I.; Yao, J.; Mollura, D.; Summers, R. M. Deep Convolutional Neur al Networks for Computer -Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning. IEEE Trans. Med. Imaging 2016, 35 (5), 1285 ...
2016
-
[23]
A Review on Deep Convolutional Neural Networks
Aloysius, N.; Geetha, M. A Review on Deep Convolutional Neural Networks. Proc. 2017 IEEE Int. Conf. Commun. Signal Process. ICCSP 2017 2018, 2018- Janua, 588 –592. https://doi.org/10.1109/ICCSP.2017.8286426
2017
-
[24]
A Framework for Designing the Architectures of Deep Convolutional Neura l Networks
Albelwi, S.; Mahmood, A. A Framework for Designing the Architectures of Deep Convolutional Neura l Networks. Entropy 2017, Vol. 19, Page 242 2017, 19 (6),
2017
-
[25]
Human Detection and Activity Classification Based on Micro -Doppler Signatures Using Deep Convolutional Neural Networks
Kim, Y.; Moon, T. Human Detection and Activity Classification Based on Micro -Doppler Signatures Using Deep Convolutional Neural Networks. IEEE Geosci. Remote Sens. Lett. 2016, 13 (1], 8 –12. https://doi.org/10.1109/LGRS.2015.2491329
2016
-
[26]
Parametric Exponential Linear Unit for Deep Convolutional Neural Networks
Trottier, L.; Gigure, P.; Chaib -Draa, B. Parametric Exponential Linear Unit for Deep Convolutional Neural Networks. Proc. - 16th IEEE Int. Conf. Mach. Learn. Appl. ICMLA 2 017 2017, 2017-Decem, 207 –214. https://doi.org/10.1109/ICMLA.2017.00038
2017
- [27]
-
[28]
Improved Convolutional Neural Network Based on Fast Exponentially Linear Unit Activation Function
Qiumei, Z.; Dan, T.; Fenghua, W. Improved Convolutional Neural Network Based on Fast Exponentially Linear Unit Activation Function. IEEE Access 2019, 7, 151359 –151367. https://doi.org/10.1109/ACCESS.2019.2948112
2019
-
[29]
S.; Ludermir, T
Gecynalda, G. S.; Ludermir, T. B.; Lima, L. M. M. R. Comparison of New Activation Functions in Neural Network for Forecasting Financial Time Series. Neural Comput. Appl. 2011, 20 (3), 417 –439. https://doi.org/10.1007/S00521-010-0407-3/METRICS
2011 doi
-
[30]
D.; Adjailia, F.; Sincak, P
Rasamoelina, A. D.; Adjailia, F.; Sincak, P. A Review of Activation Function for Artificial Neural Network. SAMI 2020 - IEEE 18th World Symp. Appl. Mach. Intell. Informatics, Proc. 2020, 281 –286. https://doi.org/10.1109/SAMI48414.2020.9108717
2020
- [31]
-
[32]
Kuzborskij, I.; Lampert, C. H. Data -Dependent Stability of Stochastic Gradient Descent. PMLR July 2018, pp 2815–2824
2018
-
[33]
Wijnhoven, R. G. J.; De With, P. H. N. Fast Training of Object Detection Using Stochastic Gradient Descent. Proc. - Int. Conf. Pattern Recognit. 2010, 424 –427. https://doi.org/10.1109/ICPR.2010.112
2010 doi
-
[35]
Female and Male Eyes
Biza, P. Female and Male Eyes. Kaggle 2021. https://doi.org/10.34740/KAGGLE/DS/1438879
2021
-
[242]
https://doi.org/10.3390/E19060242
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.