Pith. sign in

REVIEW 4 major objections 4 minor 32 references

Generalized Single-Image-Based Morphing Attack Detection Using Deep Representations from Vision Transformer

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The paper claims a frozen ImageNet-pretrained Vision Transformer feature extractor with a linear SVM improves open-set single-image morphing attack detection on digital face images.

desk verdict First ViT-for-S-MAD paper, but the key table is duplicated and the claimed gain is within noise; fixable but not acceptable as is. read the letter →

arxiv 2501.09817 v1 pith:OHTDO4A2 submitted 2025-01-16 cs.CV cs.AI

Haoyu Zhang , Raghavendra Ramachandra , Kiran Raja , Christoph Busch This is my paper · ORCID
classification cs.CVcs.AI
keywords facemorphingattackdetectionsingle-imageVisionTransformergeneralizationcross-datasettestingbiometricssecuritydeepfeatureextractionlinearSVM
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 specific version of face-morphing defense: given one suspected face photo, detect whether it is a morphed image produced by an algorithm the detector has never seen. It argues that a frozen, ImageNet-pretrained Vision Transformer (ViT-L), whose 1024-dimensional classification token is classified by a linear SVM, generalizes across five morphing algorithms on digital images better than CNN-based transfer features, hand-crafted feature ensembles, and trained residual networks. In leave-one-out cross-dataset testing on an FRGC-based database, training on four morphing algorithms and testing on the fifth, the ViT pipeline achieves a mean detection equal-error rate (D-EER, the error at which missed attacks and false alarms balance; lower is better) of 13.63% on digital images, 1.15 percentage points lower than the best baseline (14.78%, Multi-level Deep Features). The paper reports that the same advantage does not hold for print-scan or print-scan-compressed images, where ViT representations degrade and several baselines perform better. A sympathetic reader should care because open-set generalization is the operational requirement for passport-control morphing attack detection: the next attack may come from a morphing algorithm not represented in training data.

What carries the argument

The machinery is the classification token of a frozen, ImageNet-pretrained Vision Transformer (ViT-L, with 32-by-32 input patches) used as a deep feature extractor, followed by a linear SVM for binary morph/bona-fide classification. A face image is cropped by MTCNN, resized to 384 by 384 pixels, split into 144 patches of 32 by 32 pixels, and each patch is projected to a linear embedding; a learnable classification token is prepended and 1-D sinusoidal positional embeddings are added, then 24 encoder blocks with 16-head self-attention process the sequence, and the final classification token (dimension 1024) serves as the image representation. The self-attention mechanism is the load-bearing part: unlike convolutions, it has no strong image-specific inductive bias and can integrate local and global information from low layers, which the paper argues matches the widely distributed traces of morphing. The linear SVM is chosen instead of fine-tuning a deep classifier to avoid overfitting on small-to-medium datasets.

What would settle it

Run the same ViT-SVM pipeline on morphs produced by a morphing algorithm not among the five used here, or on an independent face database with a different population and camera; if the mean cross-dataset D-EER on digital images is not lower than the best baseline, the paper's generalizability claim would be refuted.

Watch

Extended reading notes

Core claim

The central claim, as stated in the paper's conclusion, is that the pure self-attention representation from a vanilla ViT yields an improvement in the generalizability of S-MAD (single-image morphing attack detection) for digital use cases. The evidence is the statistical summary of leave-one-out cross-dataset tests within the digital image type: mean D-EER 13.63% with standard deviation 11.61, lower than all seven baselines, with Multi-level Deep Features the closest at 14.78%. The paper interprets morphing traces as widely distributed across the face region, so a model that integrates local and global information from early layers should capture them more robustly than architectures with strong image-specific inductive biases. It also shows, via t-SNE visualization, that bona fide features separate well from StyleGAN-IWBF, MIPGAN-I, and MIPGAN-II morphs, while Landmark-II morphs overlap with bona fide features, consistent with Landmark-II being the hardest attack to generalize. On print-scan and print-scan-compressed inputs the ViT pipeline underperforms several baselines, which the paper attributes to the model being pretrained only on digital images and to the lower resolution of those inputs.

Load-bearing premise

The generalizability claim rests on the assumption that leave-one-out testing across five morphing algorithms on a single face database with fixed cropping and resizing is a valid proxy for the open-set conditions a passport-control detector will meet in operation, where population, camera, and post-processing can also change.

Editorial extensions

If this is right

  • A frozen ImageNet-pretrained ViT can be used as a drop-in feature extractor for single-image morphing attack detection on digital face images, without task-specific pretraining or fine-tuning.
  • The detector should generalize across morphing algorithms of both landmark-based and GAN-based families on digital inputs, not just the algorithm it was trained on.
  • Landmark-II morphs are the hardest to separate from bona fide images, so any improvement on that attack type is the most meaningful for overall robustness.
  • For print-scan and print-scan-compression pipelines, the ViT representation alone is not sufficient; the paper's results show it underperforming the Multi-modality and Multi-level Deep Features baselines.
  • Using reconstructed bona fide images in the GAN-based training sets means the classifier is aimed at morphing artifacts rather than at generic GAN image statistics, which is what allows the generalization claim to be about morphing rather than image source.

Reading between the lines

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

  • The frozen-extractor recipe likely transfers to other forensic classification tasks where training data are scarce and the artifact is spatially distributed, because the pipeline makes no task-specific image assumptions beyond the pretraining domain.
  • A testable extension is to fine-tune the last ViT blocks or fuse ViT features with the Multi-modality approach; the paper itself mentions fusion as future work, and its results suggest this could close the print-scan gap.
  • Because the leave-one-out protocol varies only the morphing algorithm while keeping the FRGC identity pool, camera, and ICAO-style processing fixed, the reported digital advantage may not survive operational shifts in population or sensor; testing on an independent face database would settle that.
  • The paper's explanation for the print-scan degradation predicts that pretraining or adapting the ViT on print-scanned data should recover much of the lost performance, which is directly testable.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper proposes a single-image morphing attack detection (S-MAD) method that extracts deep representations using a fixed, ImageNet-pretrained Vision Transformer (ViT-L) classification token and classifies them with a linear SVM. The method is evaluated in a leave-one-out protocol across five morphing algorithms (Landmark-I, Landmark-II, StyleGAN, MIPGAN-I, MIPGAN-II) on a database built from FRGC, under three processing types: digital, print-scan, and print-scan with compression. The authors benchmark against seven baselines and report that the proposed method achieves a lower mean D-EER on digital images than all baselines, concluding that ViT features improve generalizability of S-MAD for digital use cases.

Significance. If the reported results are correct, the paper would provide a useful empirical data point: a frozen ViT-L with a linear SVM can generalize across unseen morphing algorithms on digital face images. The experimental design is straightforward and largely reproducible, and the inclusion of multiple strong baselines is a positive feature. However, the current evidence is not sufficient to support the main claim. The apparent duplication of results in Table 3, the absence of significance testing, and the narrow scope of the evaluation mean that the paper's central conclusion is not reliably established. The method also performs markedly worse on print-scan and compressed images, so the practical contribution is limited to the digital modality.

major comments (4)
  1. [Table 3] The digital columns of Table 3 (training on Landmarks-II) show that every D-EER, BPCER@MACER=5%, and BPCER@MACER=10% value for the 'Proposed Method' is identical to the corresponding value for 'Multi-level Deep Features' in all five test-morphing blocks. For example, for test Landmarks-I the D-EER is exactly 14.92, BPCER@5 33.10, BPCER@10 22.64 for both rows; the same holds for the other four test morphs. Since two different feature extractors cannot produce identical results in all these cases, the digital entries for the Proposed Method in Table 3 are evidently erroneous, likely a copy-paste error. These entries are included in the mean D-EER statistics in Table 1 and Table 7, so the key quantitative claim (digital D-EER of 13.63% for the proposed method) is not reliable. The authors must recompute the missing results for this training condition and update all affected tables, statistics, and conclusions.
  2. [Section 4, Tables 1 and 7] The claimed advantage over the best baseline is a mean D-EER difference of 1.15 percentage points in Table 1 (13.63 vs 14.78) and 1.38 percentage points in Table 7 for inter-dataset-only (16.41 vs 17.79). No significance test is reported, and the standard deviations are 11–14 D-EER over only 20 inter-dataset trials per processing type. A difference of this size is well within the sampling noise of such a small, high-variance evaluation. The authors should perform paired statistical tests (e.g., Wilcoxon signed-rank test) on the 20 pairwise D-EER differences between the proposed method and each baseline, and report effect sizes or confidence intervals. Without this, the conclusion of a 'notable improvement' is not supported.
  3. [Sections 5 and 6, Abstract] The paper frames the problem as open-set and claims 'generalized' S-MAD, but the evaluation protocol varies only the morphing algorithm while holding the face database (FRGC, 140 subjects), acquisition pipeline, and image processing types fixed. The Limitation section acknowledges that only leave-one-out training was tested. As a result, the abstract's claim of handling 'unknown attacks in an open-set scenario' is an overstatement; the evidence supports only cross-morphing-algorithm generalization within a single database and fixed preprocessing. The claims should be qualified to state this clearly, or additional experiments on other databases are required to support the broader claim.
  4. [Section 6, Conclusion] The Conclusion states that 'overall detection accuracy gain' can be noted in cross-dataset testing, but Table 1 shows the proposed method has a substantially higher mean D-EER than the best baselines for print-scan (18.33% vs 7.63%) and print-scan with compression (19.09% vs 13.38%). The Limitation section also states intra-dataset accuracy is 'less or equal' to the other algorithms. The only domain where the method shows an advantage is digital inter-dataset testing. The wording of the conclusion should be restricted to that setting, otherwise it misrepresents the overall results.
minor comments (4)
  1. [Conclusion] The Conclusion says the method is 'benchmarked against two selected SOTA algorithms', but the paper clearly evaluates seven baselines. This appears to be a typo that should be corrected.
  2. [References] Reference [22] has a typo in the title: 'Morph deterction' should be 'Morph detection'. Also, references [10] and [11] are the same paper (same title, nearly the same author list) listed twice with different years; one should be removed or the citations harmonized.
  3. [Tables and text] There are several minor typographical issues: 'In this paper ,' in the abstract has an extra space before the comma, 'P .S. with Compression' in Table 7 uses an unusual abbreviation, and 'Motimodality' in the Conclusion should be 'Multi-modality'.
  4. [Section 2] The paper does not specify the SVM regularization parameter C, the SVM implementation, or whether the classification token features are normalized before classification. Providing these implementation details would improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claim is an empirical benchmark with an external pretrained feature extractor, not a derivation from its inputs.

full rationale

The paper makes an empirical claim (frozen ImageNet-pretrained ViT-L representation plus linear SVM improves cross-morphing-algorithm generalization on digital face images) and supports it with held-out D-EER measurements. There is no mathematical derivation in which the target quantity is built into the input: the ViT features are obtained from an externally trained ImageNet model, the SVM is trained on bona fide and morph labels, and the tested target morphing algorithms are left out of that training. The comparisons against CNN-based and handcrafted baselines do not depend on the authors' prior work for their validity, and the cited prior work supplies baseline methods and morphing generators rather than the paper's conclusion. The acknowledged limitations (leave-one-out training only, intra-dataset accuracy 'less or equal') and the possible Table 3 duplication issue raised by the reviewer concern statistical robustness and correctness of reported numbers, not circularity of the reasoning. No fitted parameter is renamed as a prediction, and no uniqueness theorem or self-citation is invoked to force the ViT choice. Accordingly, this is a self-contained empirical evaluation with no circularity burden.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The paper contributes an architecture choice, not a theory. Its central claim rests on empirical transfer assumptions and an undisclosed SVM hyperparameter. No new physical or conceptual entities are introduced.

free parameters (1)
  • SVM regularization parameter C = not reported
    The linear SVM classifier is trained on ViT features; the regularization parameter is not disclosed, and it can materially affect D-EER on small datasets.
assumptions (4)
  • domain assumption Morphing artifacts are widely distributed across the face, so a large receptive field from global self-attention is beneficial.
    Section 1 asserts this to motivate ViT; if artifacts are localized, ViT's advantage may vanish.
  • domain assumption Features learned by an ImageNet-pretrained ViT transfer to morphing artifact detection without fine-tuning.
    Section 2 extracts frozen CLS tokens; transfer is assumed, not demonstrated with ablation.
  • domain assumption The FRGC-based simulated morph database with five generators and three processing types is representative of operational passport morphing attacks.
    Section 3 constructs the database; generalizability claims depend on this representativeness.
  • domain assumption Reconstructed bona fide images for GAN-based morphs remove GAN-specific bias from the classifier.
    Section 3 states this design choice, which is necessary for the classifier to learn morphing traces rather than GAN artifacts.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generalized Single-Image-Based Morphing Attack Detection Using Deep Representations from Vision Transformer." pith.science (2026). https://pith.science/paper/OHTDO4A2

@misc{pith2026250109817,
  author       = {Pith},
  title        = {Pith review of: Generalized Single-Image-Based Morphing Attack Detection Using Deep Representations from Vision Transformer},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OHTDO4A2}},
  note         = {Machine review of arXiv:2501.09817}
}
read the original abstract

Face morphing attacks have posed severe threats to Face Recognition Systems (FRS), which are operated in border control and passport issuance use cases. Correspondingly, morphing attack detection algorithms (MAD) are needed to defend against such attacks. MAD approaches must be robust enough to handle unknown attacks in an open-set scenario where attacks can originate from various morphing generation algorithms, post-processing and the diversity of printers/scanners. The problem of generalization is further pronounced when the detection has to be made on a single suspected image. In this paper, we propose a generalized single-image-based MAD (S-MAD) algorithm by learning the encoding from Vision Transformer (ViT) architecture. Compared to CNN-based architectures, ViT model has the advantage on integrating local and global information and hence can be suitable to detect the morphing traces widely distributed among the face region. Extensive experiments are carried out on face morphing datasets generated using publicly available FRGC face datasets. Several state-of-the-art (SOTA) MAD algorithms, including representative ones that have been publicly evaluated, have been selected and benchmarked with our ViT-based approach. Obtained results demonstrate the improved detection performance of the proposed S-MAD method on inter-dataset testing (when different data is used for training and testing) and comparable performance on intra-dataset testing (when the same data is used for training and testing) experimental protocol.

Figures

Figures reproduced from arXiv: 2501.09817 by the authors.

Figure 1
Figure 1. Hypothesised illustration of S-MAD as open-set problem: A model trained on known morphing attacks may fail at unknown [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of our proposed method using pretrained Vision Transformer model. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Boxplot of the statistical analysis on D-EER computed for all cross-dataset testing results on FRGC morph database. (a) Ensemble [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: T-SNE plot of the feature space used in proposed method [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: T-SNE plot of the feature space used in proposed method [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: T-SNE plot of the feature space used in proposed method [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 29 canonical work pages

  1. [1]

    An image is worth 16x16 words: Trans- formers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, et al. An image is worth 16x16 words: Trans- formers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020. 2, 3

    arXiv 2010

  2. [2]

    The magic passport

    Matteo Ferrara, Annalisa Franco, and Davide Maltoni. The magic passport. In IEEE International Joint Conference on Biometrics, pages 1–7. IEEE, 2014. 1

    work page 2014

  3. [3]

    De- coupling texture blending and shape warping in face mor- phing

    Matteo Ferrara, Annalisa Franco, and Davide Maltoni. De- coupling texture blending and shape warping in face mor- phing. In 2019 International Conference of the Biometrics Special Interest Group (BIOSIG), pages 1–5. IEEE, 2019. 4, 11, 12, 13, 14, 15

    work page 2019

  4. [4]

    Face morphing detection in the presence of printing/scanning and heterogeneous image sources

    Matteo Ferrara, Annalisa Franco, and Davide Mal- toni. Face morphing detection in the presence of print- ing/scanning and heterogeneous image sources. arXiv preprint arXiv:1901.08811, 2019. 2

    work page Pith review arXiv 1901

  5. [5]

    Machine readable passports – part 9 – deployment of biometric identification and electronic storage of data in eMRTDs, 2021

    International Civil Aviation Organization. Machine readable passports – part 9 – deployment of biometric identification and electronic storage of data in eMRTDs, 2021. 4

    work page 2021

  6. [6]

    ISO/IEC CD 20059.2 Methodologies to evaluate the resistance of biometric recog- nition systems to morphing attacks

    ISO/IEC JTC1 SC37 Biometrics. ISO/IEC CD 20059.2 Methodologies to evaluate the resistance of biometric recog- nition systems to morphing attacks . International Organiza- tion for Standardization, 2023. 4

    work page 2023

  7. [7]

    Handbook of face recognition

    Anil K Jain and Stan Z Li. Handbook of face recognition . Springer, 2011. 1

    work page 2011

  8. [8]

    Face Analysis Technology Evaluation (F ATE) Part 4: MORPH - Performance of Automated Face Morph Detec- tion: Morph-performance of automated face morph detec- tion

    Mei Ngan, Patrick Grother, Kayee Hanaoka, and Jason Kuo. Face Analysis Technology Evaluation (F ATE) Part 4: MORPH - Performance of Automated Face Morph Detec- tion: Morph-performance of automated face morph detec- tion. US Department of Commerce, National Institute of Standards and Technology, 2024. 2, 4, 7

    work page 2024

Show all 32 references
  1. [9]

    P. J. Phillips, P. J. Flynn, T. Scruggs, K. W. Bowyer, Jin Chang, K. Hoffman, J. Marques, Jaesik Min, and W. Worek. Overview of the face recognition grand challenge. In 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05) , pages 947–954 vol. 1,

    2005

  2. [10]

    Morphing attack detection-database, eval- uation platform, and benchmarking

    Kiran Raja, Matteo Ferrara, Annalisa Franco, Luuk Spreeuwers, Ilias Batskos, Florens de Wit, Marta Gomez- Barrero, Ulrich Scherhag, Daniel Fischer, Sushma Krupa Venkatesh, et al. Morphing attack detection-database, eval- uation platform, and benchmarking. IEEE transactions on ...

    2020

  3. [11]

    Mor- phing attack detection-database, evaluation platform, and benchmarking

    Kiran Raja, Matteo Ferrara, Annalisa Franco, Luuk Spreeuwers, Ilias Batskos, Florens de Wit, Marta Gomez- Barrero, Ulrich Scherhag, Daniel Fischer, Sushma Krupa Venkatesh, Jag Mohan Singh, Guoqiang Li, Lo ¨ıc Berg- eron, Sergey Isadskiy, Raghavendra Ramachandra, Christian Rath...

    2021

  4. [12]

    Towards generalized morphing attack detection by learning residuals

    Kiran Raja, Gourav Gupta, Sushma Venkatesh, Raghavendra Ramachandra, and Christoph Busch. Towards generalized morphing attack detection by learning residuals. Image and Vision Computing, 126:104535, 2022. 2, 4, 5, 11, 12, 13, 14, 15, 16

    2022

  5. [13]

    Multimodality for reliable single image based face morphing attack detec- tion

    Raghavendra Ramachandra and Guoqiang Li. Multimodality for reliable single image based face morphing attack detec- tion. IEEE Access, 10:82418–82433, 2022. 4, 5, 7, 11, 12, 13, 14, 15, 16

    2022

  6. [14]

    Face morphing versus face averaging: Vulnerability and detection

    Raghavendra Ramachandra, Kiran B Raja, Sushma Venkatesh, and Christoph Busch. Face morphing versus face averaging: Vulnerability and detection. In IEEE International Joint Conference on Biometrics (IJCB) , pages 555–563, 2017. 4, 11, 12, 13, 14, 15

    2017

  7. [15]

    Raja, Sushma Venkatesh, and Christoph Busch

    Raghavendra Ramachandra, Kiran B. Raja, Sushma Venkatesh, and Christoph Busch. Transferable deep-cnn fea- tures for detecting digital and print-scanned morphed face images. In 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 1822– 1830, ...

    2017

  8. [16]

    Towards making morphing attack de- tection robust using hybrid scale-space colour texture fea- tures

    Raghavendra Ramachandra, Sushma Venkatesh, Kiran Raja, and Christoph Busch. Towards making morphing attack de- tection robust using hybrid scale-space colour texture fea- tures. In 2019 IEEE 5th International Conference on Iden- tity, Security, and Behavior Analysis (ISBA) , p...

    2019

  9. [17]

    Detecting face morphing attacks with collaborative representation of steerable features

    Raghavendra Ramachandra, Sushma Venkatesh, Kiran Raja, and Christoph Busch. Detecting face morphing attacks with collaborative representation of steerable features. In Pro- ceedings of 3rd International Conference on Computer Vi- sion and Image Processing: CVIP 2018, V olume 1...

    2018

  10. [18]

    Detecting face morphing attacks with collaborative representation of steerable features

    Raghavendra Ramachandra, Sushma Venkatesh, Kiran Raja, and Christoph Busch. Detecting face morphing attacks with collaborative representation of steerable features. In Pro- ceedings of 3rd International Conference on Computer Vi- sion and Image Processing , pages 255–265. Spri...

    2020

  11. [19]

    Imagenet-21k pretraining for the masses

    Tal Ridnik, Emanuel Ben-Baruch, Asaf Noy, and Lihi Zelnik-Manor. Imagenet-21k pretraining for the masses. arXiv preprint arXiv:2104.10972, 2021. 3

    2021 arXiv

  12. [20]

    Imagenet large scale visual recognition challenge

    Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, San- jeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, et al. Imagenet large scale visual recognition challenge. International journal of computer vision, 115:211–252, 2015. 3

    2015

  13. [21]

    Biometric systems under morphing attacks: Assessment of morphing techniques and vulnerability reporting

    Ulrich Scherhag, Andreas Nautsch, Christian Rathgeb, Marta Gomez-Barrero, Raymond NJ Veldhuis, Luuk Spreeuwers, Maikel Schils, Davide Maltoni, Patrick Grother, Sebastien Marcel, Breithaupt Ralph, Raghavendra Ra- machandra, and Christoph Busch. Biometric systems under morphing ...

    2017

  14. [22]

    Morph deterction from single face image: A multi-algorithm fusion approach

    Ulrich Scherhag, Christian Rathgeb, and Christoph Busch. Morph deterction from single face image: A multi-algorithm fusion approach. In Proceedings of the 2018 2nd Interna- tional Conference on Biometric Engineering and Applica- tions, pages 6–12, 2018. 2

    2018

  15. [23]

    Style your face morph and improve your face morphing attack de- tector

    Clemens Seibold, Anna Hilsmann, and Peter Eisert. Style your face morph and improve your face morphing attack de- tector. In 2019 International Conference of the Biometrics Special Interest Group (BIOSIG), pages 1–6. IEEE, 2019. 2

    2019

  16. [24]

    Visualizing data using t-sne

    Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-sne. Journal of machine learning research , 9 (11), 2008. 5

    2008

  17. [25]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017. 2

    2017

  18. [26]

    Multilevel fusion of deep features for reliable single image based face morphing attack de- tection

    Sushma Venkatesh. Multilevel fusion of deep features for reliable single image based face morphing attack de- tection. In 2022 International Conference on Electrical, Computer , Communications and Mechatronics Engineering (ICECCME). IEEE, 2022. 2, 4, 5, 11, 12, 13, 14, 15, 16

    2022

  19. [27]

    Single image face morphing attack detection using ensemble of features

    Sushma Venkatesh, Raghavendra Ramachandra, Kiran Raja, and Christoph Busch. Single image face morphing attack detection using ensemble of features. In 23rd International Conference on Information Fusion, pages 1–5, 2020. 2, 4, 5, 11, 12, 13, 14, 15, 16

    2020

  20. [28]

    Can gan generated morphs threaten face recognition systems equally as landmark based morphs? - vulnerability and detection

    Sushma Venkatesh, Haoyu Zhang, Raghavendra Ramachan- dra, Kiran Raja, Naser Damer, and Christoph Busch. Can gan generated morphs threaten face recognition systems equally as landmark based morphs? - vulnerability and detection. In 2020 International Workshop on Biometrics and ...

    2020

  21. [29]

    Face morphing attack generation and detection: A comprehensive survey

    Sushma Venkatesh, Raghavendra Ramachandra, Kiran Raja, and Christoph Busch. Face morphing attack generation and detection: A comprehensive survey. IEEE Transactions on Technology and Society, 2(3):128–145, 2021. 1, 2

    2021

  22. [30]

    Mip- gan—generating strong and high quality morphing attacks using identity prior driven gan

    Haoyu Zhang, Sushma Venkatesh, Raghavendra Ramachan- dra, Kiran Raja, Naser Damer, and Christoph Busch. Mip- gan—generating strong and high quality morphing attacks using identity prior driven gan. IEEE Transactions on Bio- metrics, Behavior , and Identity Science, 3(3):365–38...

    2021

  23. [31]

    Joint face detection and alignment using multitask cascaded convolutional networks

    Kaipeng Zhang, Zhanpeng Zhang, Zhifeng Li, and Yu Qiao. Joint face detection and alignment using multitask cascaded convolutional networks. IEEE signal processing letters , 23 (10):1499–1503, 2016. 2 9 A. Appendix A.1. Complete tables of MAD performances In this appendix, deta...

    2016

  24. [2019]

    2, 4, 5, 7, 11, 12, 13, 14, 15, 16

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.