Pith. sign in

REVIEW 5 major objections 4 minor 39 references

LitMAS: A Lightweight and Generalized Multi-Modal Anti-Spoofing Framework for Biometric Security

T0 review · 5 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A 6M-parameter model unifies anti-spoofing across four biometric modalities, beating seven modality-specific baselines on average error rate.

desk verdict A plausible unified anti-spoofing model whose headline SOTA claim rests on a comparison protocol that is not clearly fair and an averaging procedure that is not fully defined. read the letter →

arxiv 2506.06759 v1 pith:IQTZDRHI submitted 2025-06-07 cs.CV

Nidheesh Gorthi , Kartik Thakral , Rishabh Ranjan , Richa Singh , Mayank Vatsa This is my paper
classification cs.CV
keywords anti-spoofingpresentationattackdetectionmulti-modalbiometricslightweightmodelDeiT-Tinymodality-alignedconcentrationlossmixtureofprojectionexpertsequalerrorrate
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

This paper claims that a single lightweight network can detect presentation attacks across four biometric modalities—speech, face, iris, and fingerprint—using one shared transformer backbone. The proposed LitMAS model, at roughly six million parameters, is trained jointly on seven datasets and reported to reach a 4.86% average equal error rate, outperforming seven modality-specific baselines retrained under the same unified protocol by 1.36 percentage points. If the result holds, edge devices could run one anti-spoofing model instead of a suite of modality-specialized ones, a nd the framework offers a common protocol for comparing cross-modal spoof detection.

What carries the argument

The load-bearing component is the Modality-Aligned Concentration (MAC) loss, a per-modality softmax-concentration objective whose denominator includes bonafide samples of that modality and spoof samples of all modalities, so each training step pulls genuine embeddings toward their modality center and pushes every spoof away. On top sits the Mixture of Projection Experts (MoPE), a set of modality-specific projection heads that map the shared DeiT-Tiny [CLS] embedding (dimension 192 in Step 1) to a 512-dimensional space, capturing cues such as replay artifacts in speech or texture anomalies in fingerprints before a single classifier outputs the spoof score.

What would settle it

Reproduce LitMAS with a publicly specified speech front-end, for example 64-band log-mel spectrograms with 25 ms windows and 10 ms hop, and check whether the ASVSpoof-2019 EER stays near 4.54% and the seven-dataset average near 4.86%. If those numbers move materially or fall behind a unimodal speech countermeasure trained under its native protocol, the claimed cross-modal gain is an artifact of the audio transform; likewise, an independent re-implementation of the Table 5 ablation (MAC and MoPE both off vs both on) either reproduces the 6.17% to 4.86% improvement or falsifies the synergy claim.

Watch

Extended reading notes

Core claim

LitMAS is a two-stage framework: first, a Modality-Aligned Concentration (MAC) loss shapes a shared embedding space by pulling bonafide samples of each modality toward that modality's center while repelling spoof samples of every modality; second, a Mixture of Projection Experts (MoPE) layer projects embeddings through per-modality heads before a lightweight classifier makes the live-versus-spoof decision. The paper reports that combining both stages yields an average AUC of 0.9902 and an average EER of 4.86% across ASVSpoof-2019, Iris CSD, LivDet-2017, MSU-MFSD, Replay Attack, Silicone Mask, and MLFP, with the full system outperforming each of seven baselines on average while using only about six million parameters. The ablation study shows that MAC pre-training or MoPE alone improves the average EER only to roughly 5.7%, whereas the two together reach 4.86%.

Load-bearing premise

All four biometric modalities can be reduced to a fixed three-channel image-like input that a DeiT-Tiny transformer can ingest, but the paper never specifies how speech is converted into that input, so the reported speech error rate and the fair comparison with speech baselines rest on an unstated preprocessing choice.

Editorial extensions

If this is right

  • A single ~6M-parameter model can replace modality-specific anti-spoofing systems on devices with tight memory and compute budgets.
  • The unified training protocol gives a direct way to compare speech, face, iris, and fingerprint anti-spoofing methods on the same footing, since all baselines are retrained on the combined multi-modal set.
  • The MAC loss pre-training step could be reused as a generic initialization for other cross-modal classification tasks where real-class geometry matters.
  • Because the model processes one modality at a time, it can be dropped into existing single-modality pipelines without changing the sensor setup.

Reading between the lines

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

  • The speech leg of the evaluation is the least specified: the paper never states the audio representation (spectrogram type, size, resampling, normalization) fed to DeiT-Tiny, so the 4.54% speech EER is reproducible only if that choice is inferred.
  • The paper compares against baselines trained on the combined multi-modal data; a unimodal specialist trained only on its own dataset and protocol might still beat LitMAS on that modality, and that comparison is not reported.
  • MAC loss could be seen as a variant of prototype-based metric learning; a testable extension is replacing per-modality bonafide centers with a single shared center to see whether cross-modal generalization improves or degrades.
  • The framework treats each biometric trait independently; true multi-modal fusion where several traits are presented together is left open.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 4 minor

Summary. The paper proposes LitMAS, a unified anti-spoofing framework that maps speech, face, iris, and fingerprint inputs into a single DeiT-Tiny backbone with a Modality-Aligned Concentration (MAC) loss and a Mixture of Projection Experts (MoPE) head. The model is pre-trained with the MAC loss to pull bonafide embeddings of each modality toward modality centers and push spoof embeddings away, then fine-tuned with modality-specific projection experts and a lightweight classifier. The authors train on a combined multi-modal training set, evaluate per modality on ASVSpoof, Iris CSD, LivDet-2017, MSU-MFSD, Replay Attack, Silicone Mask, and MLFP, and report that LitMAS achieves an average EER of 4.86% with about 6M parameters, outperforming retrained baselines by 1.36%. The paper includes AUC/EER tables, min t-DCF/BPCER@APCER=1% tables, per-dataset APCER/BPCER, an ablation study, and t-SNE visualizations. Code and trained models are stated to be publicly available.

Significance. If the reported results are reproducible and the comparison protocol is fair, LitMAS would be a practically valuable contribution: a single lightweight model covering four biometric modalities is more deployable on edge devices than modality-specific countermeasures, and the MAC loss plus MoPE combination is a reasonable design for cross-modal spoof detection. The paper also deserves credit for targeting seven datasets, releasing code, and including ablation and visualization analyses. However, the central quantitative claims are currently not supported by the evidence as presented: the headline average EER is not consistently defined, the baseline retraining protocol may disadvantage modality-native methods, the audio input representation is unspecified, and all results are single-run with no uncertainty quantification. These issues are load-bearing because the paper's main selling point is the numerical superiority of LitMAS over prior state of the art.

major comments (5)
  1. [Abstract and Tables 1, 4] The claim of "average EER across seven datasets" is numerically inconsistent with the paper's own tables. Table 1 reports EERs for four modality-pooled groups, whose mean is (4.54 + 3.13 + 5.06 + 6.70)/4 = 4.86, but the paper says this is across seven datasets. If one instead averages the per-dataset EERs implied by Table 4 (where APCER and BPCER at the EER threshold are approximately equal), the mean is about 7.15%, not 4.86%. The aggregation protocol for the seven datasets into the four modality entries is never defined, so the headline 4.86% figure and the 1.36% improvement over state of the art are not reproducible from the reported results.
  2. [§3 Baseline Methods] The baseline comparison is not a fair test of state-of-the-art performance. The paper states that "all baselines are trained on the combined multi-modal dataset," with input layers altered (e.g., SpecRNet modified to accept 3-channel inputs and DeiT-Tiny used as the backbone for AA ViT). This unified protocol forces modality-native methods such as LCNN and SpecRNet to ingest face, iris, and fingerprint images, and forces fingerprint-specific MoSFPAD to process speech and face inputs. The reported margins over these retrained, possibly suboptimal baselines therefore do not establish superiority over published per-dataset state of the art. The authors should additionally report per-dataset comparisons using the baselines' native protocols, or at least provide full retraining details (hyperparameters, input preprocessing per modality, and seeds) so the reader can judge whether the baselines were tuned fairly.
  3. [§3 Implementation details] The audio input representation for LitMAS is never specified. Since every input is fed to DeiT-Tiny as a fixed image-like 3-channel tensor, the ASVSpoof speech signals must be converted to some spectrogram or other image representation, but the paper does not state the spectrogram type, size, windowing, resampling rate, or normalization. This choice is load-bearing for the speech EER of 4.54% and for the fair comparison with speech-specific baselines LCNN and SpecRNet, which use their own audio front-ends. Without this information the speech experiments are not reproducible.
  4. [Table 5 and §4] All results appear to be from a single training run, with no error bars, confidence intervals, or significance tests. In the ablation study, adding MAC alone or MoPE alone changes EER from 6.17% to 5.64% and 5.75%, respectively, a difference of roughly 0.4-0.5 percentage points; the difference between the two single-component configurations is only 0.11 percentage points. Given typical run-to-run variance in training Transformers on small and imbalanced datasets, the claim that both components are synergistic and that the final 4.86% is a genuine improvement requires multiple seeds and a report of mean and variance. As it stands, the ablative evidence is not statistically persuasive.
  5. [Eq. (2)] The MAC loss is not fully specified. The term y_hat_i is described as the "softmax-transformed label," but no formula or procedure is given for computing it from the binary bonafide/spoof label, nor is it clear how the softmax over the batch of cosine similarities in Eq. (2) implements the intended pull-and-push behavior. Since the MAC loss is a central contribution, this missing definition prevents the reader from reimplementing or theoretically checking the method.
minor comments (4)
  1. [§3 / Reference [26]] Reference [26] is cited as "ASVSpoof-2019 [26]" in Table 3, but the reference itself is the ASVspoof 2015 paper; please correct the citation or the dataset label.
  2. [Table 2 caption] The caption spells the baseline as "MosFPAD" while the text and other tables use "MoSFPAD"; please standardize the spelling.
  3. [§2 / MoPE] The projection dimension k is described only as "possibly larger" than d; please specify the exact value used in the experiments (the text says [CLS] features of size 192 are projected to 512, but this should be stated explicitly in the MoPE definition).
  4. [Table 1] The "Average" column is computed over the four modality-pooled rows, not over seven datasets; please add a caption note explaining this, and consider reporting per-dataset EERs so the claim "across seven datasets" can be verified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: LitMAS's test-set EER results are held-out evaluations on external benchmarks, and no equation or fitted parameter reduces to the claimed outcome.

full rationale

The paper is an empirical machine-learning study rather than a derivation. The central claim (average EER of 4.86% with ~6M parameters) is obtained by training LitMAS on a combined multi-modal training set (D_train = union of D_train^(m), Eq. 1) and evaluating on held-out test samples from seven external datasets (Table 3 lists 176,875 test samples). The MAC loss (Eqs. 2-3) uses modality centers computed from bonafide embeddings; while the centers depend on the training distribution, this is standard prototype-based representation learning and does not by construction determine test EER. The MoPE layer selects a projection head by modality label and is trained with cross-entropy, another standard supervised component. No result is imported from a self-citation chain: the DeiT-Tiny backbone is cited to Touvron et al. [33], an external source, and the self-citations in the references concern other datasets or prior deepfake/PAD work that is not load-bearing for the reported numbers. Concerns that baselines were re-trained on a unified protocol and that the 'seven dataset' average is not explicitly aggregated are comparison-protocol and reproducibility issues, not circularity: the reported improvement does not reduce by definition to a fitted parameter or to an assumption of the method.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

The paper introduces no fitted scalar constants and no new physical or conceptual entities. The central claim depends on the architectural assumptions listed above, especially the undocumented audio preprocessing and the undefined label transform in the loss.

assumptions (2)
  • domain assumption A single DeiT-Tiny backbone can represent audio, face, iris, and fingerprint inputs in a shared embedding space.
    Section 3 (Implementation details) adopts DeiT-Tiny for all modalities without specifying how audio is transformed into the model's expected image-like input.
  • ad hoc to paper The softmax-transformed label y-hat_i in Eq. (2) implements the intended pull-and-push behavior for bonafide and spoof samples.
    Eq. (2) defines the MAC loss in terms of y-hat_i but never defines it; the claimed push-away effect for spoof samples depends on this undefined label transform.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LitMAS: A Lightweight and Generalized Multi-Modal Anti-Spoofing Framework for Biometric Security." pith.science (2026). https://pith.science/paper/IQTZDRHI

@misc{pith2026250606759,
  author       = {Pith},
  title        = {Pith review of: LitMAS: A Lightweight and Generalized Multi-Modal Anti-Spoofing Framework for Biometric Security},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IQTZDRHI}},
  note         = {Machine review of arXiv:2506.06759}
}
abstract

Biometric authentication systems are increasingly being deployed in critical applications, but they remain susceptible to spoofing. Since most of the research efforts focus on modality-specific anti-spoofing techniques, building a unified, resource-efficient solution across multiple biometric modalities remains a challenge. To address this, we propose LitMAS, a $\textbf{Li}$gh$\textbf{t}$ weight and generalizable $\textbf{M}$ulti-modal $\textbf{A}$nti-$\textbf{S}$poofing framework designed to detect spoofing attacks in speech, face, iris, and fingerprint-based biometric systems. At the core of LitMAS is a Modality-Aligned Concentration Loss, which enhances inter-class separability while preserving cross-modal consistency and enabling robust spoof detection across diverse biometric traits. With just 6M parameters, LitMAS surpasses state-of-the-art methods by $1.36\%$ in average EER across seven datasets, demonstrating high efficiency, strong generalizability, and suitability for edge deployment. Code and trained models are available at https://github.com/IAB-IITJ/LitMAS.

Figures

Figures reproduced from arXiv: 2506.06759 by the authors.

Figure 1
Figure 1. Comparison of (a) existing multi-modal anti-spoofing approach and (b) a unified approach (proposed). Over the years, substantial progress has been made in anti￾spoofing research across individual biometric modalities. In voice anti-spoofing, techniques such as linear frequency cep￾stral coefficients (LFCC) and constant-Q cepstral coefficients (CQCC) [6] have been widely used. At the same time, re￾cent approaches foc… view at source ↗
Figure 2
Figure 2. Illustration of the proposed LitMAS architecture (best viewed in color). The left block demonstrates the first step of the pre￾training task with the MAC loss, and the right block demonstrates the second step of MoPE architecture used for the downstream task. preserving modality-specific information. Additionally, our Mixture of Projection Experts (MoPE) architecture balances modality-invariant representations with … view at source ↗
Figure 3
Figure 3. t-SNE visualizations depicting the separation between spoofed and bonafide samples across different ablation configurations. layer of SpecRNet [24] to accept 3-channel inputs, and DeiT￾Tiny is used as the backbone for AAViT [11]. Experiments are conducted on an Nvidia A40 and three A30 GPUs (48GB and 24GB VRAM, respectively). 4. Results and Discussion Generalization on Multi-Modal Anti-Spoofing Detection [PITH_FULL… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

39 extracted references · 38 canonical work pages

  1. [1]

    LitMAS: A Lightweight and Generalized Multi-Modal Anti-Spoofing Framework for Biometric Security

    Introduction Biometric authentication has become a cornerstone of secure identity verification, widely used in banking, mobile authentica- tion, access control, and emerging technologies such as AR/VR systems and smart assistants. These systems leverage speech, face, iris, and fingerprint recognition, which can be used indi- vidually or in combination to ...

    work page Pith review arXiv 2025

  2. [2]

    LetM= {m1, m2,

    Proposed LitMAS Framework for Multi-Modal Anti-Spoofing The proposed framework is illustrated in Figure 2. LetM= {m1, m2, . . . , m|M|}be the set of biometric modalities we aim to protect against spoofing (e.g., speech, face, iris, fingerprint). Let the combined training set be defined as Dtrain = [ m∈M D(m) train (1) whereD (m) train is the training subs...

    work page arXiv

  3. [3]

    Statistics for all the utilized datasets are pre- sented in Table 3

    Experimental Setup Datasets and Pre-processing:To evaluate the generalizability of our approach, we conduct experiments on diverse datasets spanning multiple biometric modalities, covering various types of spoofing attack. Statistics for all the utilized datasets are pre- sented in Table 3. Baseline Methods:We evaluate LitMAS against state-of-the- art met...

    work page 2019

  4. [4]

    As shown,LitMAScon- sistently outperforms all baselines, achieving superior average AUC and EER

    Results and Discussion Generalization on Multi-Modal Anti-Spoofing Detection: Table 1 summarizes the detection performance of the proposed LitMASframework compared against state-of-the-art methods in terms of AUC and EER across four biometric modalities: speech, iris, face, and fingerprint. As shown,LitMAScon- sistently outperforms all baselines, achievin...

    work page

  5. [5]

    Conclusion We presentedLitMAS, a unified anti-spoofing framework that leverages a novel Modality-Aligned Concentration (MAC) loss and a Mixture of Projection Experts (MoPE) to jointly han- dle speech, face, iris, and fingerprint attacks. Through ex- tensive experiments, we demonstrated that LitMAS not only disentangles modality-specific features, but also...

    work page

  6. [6]

    Gorthi was supported by ACM IKDD Uplink Internship and Thakral by the PMRF Fellowship

    Acknowledgement This research is supported by a grant from IndiaAI and Meta via the Srijan: Centre of Excellence for Generative AI. Gorthi was supported by ACM IKDD Uplink Internship and Thakral by the PMRF Fellowship

    work page

  7. [7]

    Phygitalnet: Unified face presentation attack detection via one-class isolation learning,

    K. Thakral, S. Mittal, M. Vatsa, and R. Singh, “Phygitalnet: Unified face presentation attack detection via one-class isolation learning,” in2023 IEEE 17th International Conference on Auto- matic Face and Gesture Recognition (FG). IEEE, 2023, pp. 1–6

    work page 2023

  8. [8]

    Vl-fas: Domain generalization via vision-language model for face anti- spoofing,

    H. Fang, A. Liu, N. Jiang, Q. Lu, G. Zhao, and J. Wan, “Vl-fas: Domain generalization via vision-language model for face anti- spoofing,” inICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp. 4770–4774

    work page 2024

Show all 39 references
  1. [9]

    Audio replay attack detection using high-frequency fea- tures

    M. Witkowski, S. Kacprzak, P. Zelasko, K. Kowalczyk, and J. Galka, “Audio replay attack detection using high-frequency fea- tures.” inInterspeech, 2017, pp. 27–31

    2017

  2. [10]

    Robust iris presentation attack detection fusing 2d and 3d information,

    Z. Fang, A. Czajka, and K. W. Bowyer, “Robust iris presentation attack detection fusing 2d and 3d information,”IEEE Transac- tions on Information Forensics and Security, vol. 16, pp. 510–520, 2020

    2020

  3. [11]

    Fingerprint pre- sentation attack detector using global-local model,

    H. Liu, W. Zhang, F. Liu, H. Wu, and L. Shen, “Fingerprint pre- sentation attack detector using global-local model,”IEEE Trans- actions on Cybernetics, vol. 52, no. 11, pp. 12 315–12 328, 2022

    2022

  4. [12]

    A new feature for auto- matic speaker verification anti-spoofing: Constant q cepstral co- efficients,

    M. Todisco, H. Delgado, and N. Evans, “A new feature for auto- matic speaker verification anti-spoofing: Constant q cepstral co- efficients,” inThe Speaker and Language Recognition Workshop (Odyssey 2016), 2016, pp. 283–290

    2016

  5. [13]

    Context encoded multi-modal attention network for detecting audio spoofing,

    R. Ranjan, M. Vatsa, and R. Singh, “Context encoded multi-modal attention network for detecting audio spoofing,” in2024 IEEE In- ternational Joint Conference on Biometrics (IJCB). IEEE, 2024, pp. 1–11

    2024

  6. [14]

    Sv-deit: Speaker veri- fication with deitcap spoofing detection,

    R. Rishabh and M. Vatsa, and R. Singh, “Sv-deit: Speaker veri- fication with deitcap spoofing detection,” in2023 IEEE Interna- tional Joint Conference on Biometrics (IJCB). IEEE, 2023, pp. 1–10

    2023

  7. [15]

    Data augmen- tation techniques for physical access in voice anti-spoofing,

    J. C. Sanchez, A. M. Peinado, and A. M. Gomez, “Data augmen- tation techniques for physical access in voice anti-spoofing,” in IberSPEECH 2024, 2024, pp. 1–5

    2024

  8. [16]

    Rethinking generalizable face anti-spoofing via hierarchical prototype-guided distribution refinement in hyperbolic space,

    C. Hu, K.-Y . Zhang, T. Yao, S. Ding, and L. Ma, “Rethinking generalizable face anti-spoofing via hierarchical prototype-guided distribution refinement in hyperbolic space,” in2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 1032–1041

    2024

  9. [17]

    Adaptive-avg- pooling based attention vision transformer for face anti-spoofing,

    J. Yang, F. Chen, R. K. Das, Z. Zhu, and S. Zhang, “Adaptive-avg- pooling based attention vision transformer for face anti-spoofing,” inICASSP 2024 - 2024 IEEE International Conference on Acous- tics, Speech and Signal Processing (ICASSP), 2024, pp. 3875– 3879

    2024

  10. [18]

    Assessing the performance of efficient face anti-spoofing detection against physical and digi- tal presentation attacks,

    L. S. Luevano, Y . Mart ´ınez-D´ıaz, H. M ´endez-V´azquez, M. Gonzalez-Mendoza, and D. Frey, “Assessing the performance of efficient face anti-spoofing detection against physical and digi- tal presentation attacks,” in2024 IEEE/CVF Conference on Com- puter Vision and Pattern R...

    2024

  11. [19]

    Illusion: Unveiling truth with a comprehensive multi-modal, multi-lingual deepfake dataset,

    K. Thakral, R. Ranjan, A. Singh, A. Jain, M. Vatsa, and R. Singh, “Illusion: Unveiling truth with a comprehensive multi-modal, multi-lingual deepfake dataset,” inThe Thirteenth International Conference on Learning Representations

  12. [20]

    Df-platter: Multi-face heterogeneous deepfake dataset,

    K. Narayan, H. Agarwal, K. Thakral, S. Mittal, M. Vatsa, and R. Singh, “Df-platter: Multi-face heterogeneous deepfake dataset,” inProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, 2023, pp. 9739–9748

    2023

  13. [21]

    Deephy: On deepfake phylogeny,

    K. Narayan, H. Agarwal, K. Thakral, S. Mittal, M. Vatsa, and R. Singh, “Deephy: On deepfake phylogeny,” in2022 IEEE Interna- tional Joint Conference on Biometrics (IJCB). IEEE, 2022, pp. 1–10

    2022

  14. [22]

    Deephynet: Towards detecting phylogeny in deep- fakes,

    K. Thakral, H. Agarwal, K. Naraya, S. Mittal, M. Vatsa, and R. Singh, “Deephynet: Towards detecting phylogeny in deep- fakes,”IEEE Transactions on Biometrics, Behavior, and Identity Science, 2024

    2024

  15. [23]

    Low- quality deepfake detection via unseen artifacts,

    S. Chhabra, K. Thakral, S. Mittal, M. Vatsa, and R. Singh, “Low- quality deepfake detection via unseen artifacts,”IEEE Transac- tions on Artificial Intelligence, vol. 5, no. 4, pp. 1573–1585, 2023

    2023

  16. [24]

    Desi: Deepfake source identifier for so- cial media,

    K. Narayan, H. Agarwal, S. Mittal, K. Thakral, S. Kundu, M. Vatsa, and R. Singh, “Desi: Deepfake source identifier for so- cial media,” inProceedings of the IEEE/CVF conference on com- puter vision and pattern recognition, 2022, pp. 2858–2867

    2022

  17. [25]

    Generalized con- tact lens iris presentation attack detection,

    A. Agarwal, A. Noore, M. Vatsa, and R. Singh, “Generalized con- tact lens iris presentation attack detection,”IEEE Transactions on Biometrics, Behavior, and Identity Science, vol. 4, no. 3, pp. 373– 385, 2022

    2022

  18. [26]

    Privacy-safe iris presenta- tion attack detection,

    P. T. Mahsa Mitcheff and A. Czajka, “Privacy-safe iris presenta- tion attack detection,” inIEEE International Joint Conference on Biometrics, 2024

    2024

  19. [27]

    Vit unified: Joint fingerprint recognition and presentation attack detection,

    S. A. Grosz, K. P. Wijewardena, and A. K. Jain, “Vit unified: Joint fingerprint recognition and presentation attack detection,” in2023 IEEE International Joint Conference on Biometrics (IJCB), 2023, pp. 1–9

    2023

  20. [28]

    Mosfpad: An end-to-end ensemble of mobilenet and support vector classifier for fingerprint presentation attack detection,

    A. Rai, S. Dey, P. Patidar, and P. Rai, “Mosfpad: An end-to-end ensemble of mobilenet and support vector classifier for fingerprint presentation attack detection,”Computers & Security, vol. 148, p. 104069, 2025. [Online]. Available: https://www.sciencedirect. com/science/artic...

    2025

  21. [29]

    A comparative study on recent neu- ral spoofing countermeasures for synthetic speech detection,

    X. Wang and J. Yamagishi, “A comparative study on recent neu- ral spoofing countermeasures for synthetic speech detection,” in Interspeech 2021, 2021, pp. 4259–4263

    2021

  22. [30]

    Specrnet: Towards faster and more accessible audio deepfake detection,

    P. Kawa, M. Plata, and P. Syga, “Specrnet: Towards faster and more accessible audio deepfake detection,” in2022 IEEE Inter- national Conference on Trust, Security and Privacy in Computing and Communications (TrustCom), 2022, pp. 792–799

    2022

  23. [31]

    Iris liveness detection us- ing a cascade of dedicated deep learning networks,

    J. E. Tapia, S. Gonzalez, and C. Busch, “Iris liveness detection us- ing a cascade of dedicated deep learning networks,”IEEE Trans- actions on Information Forensics and Security, vol. 17, pp. 42–52, 2022

    2022

  24. [32]

    Asvspoof 2015: the first automatic speaker verification spoofing and countermeasures challenge,

    Z. Wu, T. Kinnunen, N. Evans, J. Yamagishi, C. Hanilc ¸i, M. Sahidullah, and A. Sizov, “Asvspoof 2015: the first automatic speaker verification spoofing and countermeasures challenge,” in Interspeech 2015, 2015, pp. 2037–2041

    2015

  25. [33]

    Detecting medley of iris spoofing attacks using desist,

    N. Kohli, D. Yadav, M. Vatsa, R. Singh, and A. Noore, “Detecting medley of iris spoofing attacks using desist,” in2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), 2016, pp. 1–6

    2016

  26. [34]

    Livdet 2017 fingerprint liveness de- tection competition 2017,

    V . Mura, G. Orr`u, R. Casula, A. Sibiriu, G. Loi, P. Tuveri, L. Ghi- ani, and G. L. Marcialis, “Livdet 2017 fingerprint liveness de- tection competition 2017,” in2018 international conference on biometrics (ICB). IEEE, 2018, pp. 297–302

    2017

  27. [35]

    Face spoof detection with image distortion analysis,

    D. Wen, H. Han, and A. K. Jain, “Face spoof detection with image distortion analysis,”IEEE Transactions on Information Forensics and Security, vol. 10, no. 4, pp. 746–761, 2015

    2015

  28. [36]

    On the effectiveness of local binary patterns in face anti-spoofing,

    I. Chingovska, A. Anjos, and S. Marcel, “On the effectiveness of local binary patterns in face anti-spoofing,” in2012 BIOSIG - Proceedings of the International Conference of Biometrics Spe- cial Interest Group (BIOSIG), 2012, pp. 1–7

    2012

  29. [37]

    De- tecting silicone mask-based presentation attack via deep dictio- nary learning,

    I. Manjani, S. Tariyal, M. Vatsa, R. Singh, and A. Majumdar, “De- tecting silicone mask-based presentation attack via deep dictio- nary learning,”IEEE Transactions on Information Forensics and Security, vol. 12, no. 7, pp. 1713–1723, 2017

    2017

  30. [38]

    Face presentation attack with latex masks in multi- spectral videos,

    A. Agarwal, D. Yadav, N. Kohli, R. Singh, M. Vatsa, and A. Noore, “Face presentation attack with latex masks in multi- spectral videos,” in2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2017, pp. 275– 283

    2017

  31. [39]

    Training data-efficient image transformers & distilla- tion through attention,

    H. Touvron, M. Cord, M. Douze, F. Massa, A. Sablayrolles, and H. J´egou, “Training data-efficient image transformers & distilla- tion through attention,” inInternational conference on machine learning. PMLR, 2021, pp. 10 347–10 357

    2021

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.