pith. sign in

arxiv: 2606.12074 · v1 · pith:GFSE4LWZnew · submitted 2026-06-10 · 💻 cs.CV · cs.AI· eess.IV

Non-frontal face recognition using GANs and memristor-based classifiers

Semih Vazgecen , Cristian Sestito , Spyros Stathopoulos , Themis Prodromakis This is my paper

Pith reviewed 2026-06-27 10:10 UTC · model grok-4.3

classification 💻 cs.CV cs.AIeess.IV
keywords face recognitionGANmemristorneuromorphicpose variationfrontalizationedge computing
0
0 comments X

The pith

A lightweight GAN frontalizes non-frontal faces before memristor hardware performs the recognition step.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets out to show that non-frontal pose changes can be handled by first passing images through a generative adversarial network that produces frontal views and then routing those views to memristor-based classifiers for identification. A reader would care because standard deep-learning face systems demand too much power and memory for devices such as drones that must work in real time. The reported experiments reach identification accuracies up to 96 percent on two separate datasets when the two components operate together. The approach therefore claims to cut computational cost while still coping with the main source of error in unconstrained face imagery. If the claim holds, it supplies one concrete route for moving face recognition onto low-power neuromorphic chips without first solving every pose problem in software alone.

Core claim

The authors claim that pairing a lightweight GAN for pose frontalisation with memristor-based neuromorphic classifiers produces a working face-recognition pipeline that maintains up to 96 percent identification accuracy on two datasets while remaining suitable for resource-constrained platforms.

What carries the argument

The integration of a lightweight GAN that converts non-frontal images into frontal views with subsequent classification performed by memristor-based neuromorphic hardware.

If this is right

  • Face recognition becomes feasible on drones and other edge devices that cannot run full deep-learning models.
  • The computational load of handling pose variation is shifted partly to a generative front-end that can be kept lightweight.
  • Memristive neuromorphic chips can be used directly for identification once a pose-normalised image is supplied.
  • The same pairing of adversarial normalisation and memristor classification could be applied to other real-world distortions if the GAN is retrained.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the GAN output contains subtle artefacts, those artefacts may interact with memristor variability in ways that standard software classifiers would not reveal.
  • The framework could be extended to handle simultaneous pose and illumination changes by training the GAN on multi-attribute frontalisation targets.
  • Hardware-in-the-loop testing on actual memristor arrays would be needed to confirm that the reported accuracy survives device non-idealities not captured in simulation.

Load-bearing premise

Memristor classifiers continue to deliver high accuracy on the synthetic frontal images produced by the GAN rather than suffering large drops from domain shift or device noise.

What would settle it

Run the memristor hardware on the same set of subjects once with real frontal photographs and once with GAN-generated frontal versions of angled photographs; a substantial accuracy gap between the two runs would falsify the claim.

read the original abstract

Face recognition systems have advanced significantly through deep learning techniques, delivering high performance and robustness in complex scenarios. However, these approaches incur substantial computational overhead, limiting their in situ applicability in resource-constrained platforms such as drones, where they can address challenges including non-frontal facial imagery. Memristor-based neuromorphic systems have emerged as a compelling approach for edge AI applications, combining biologically inspired processing with efficient and scalable computation. In this work, we propose a facial recognition framework that addresses non-frontal pose variations by integrating lightweight generative adversarial network (GAN)-based pose frontalisation with memristor-based neuromorphic recognition. The experimental results on two datasets demonstrate the effectiveness of combining adversarial learning with memristive technology, achieving up to 96% identification accuracy. The proposed approach alleviates the computational bottlenecks of conventional AI and offers a scalable, efficient solution for face recognition in dynamic real-world environments.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a facial recognition framework integrating lightweight GAN-based pose frontalization with memristor-based neuromorphic classifiers to handle non-frontal pose variations in resource-constrained settings such as drones, reporting up to 96% identification accuracy on two datasets.

Significance. If the integration claim holds under proper validation, the work would illustrate a route to low-power, in-situ face recognition by pairing adversarial synthesis with analog neuromorphic hardware, addressing the computational cost of conventional deep-learning pipelines.

major comments (2)
  1. [Abstract] Abstract: the central claim of 'up to 96% identification accuracy' is asserted without any description of the two datasets, baselines, training protocols, error bars, or hardware implementation, supplying no evidence that the claim is supported.
  2. [Proposed framework] Proposed framework (integration section): the claim that memristor-based classifiers retain high accuracy on GAN-synthesized frontal faces rests on the untested assumptions that (1) GAN outputs lie in the same distribution as real frontal training data and (2) memristor non-idealities (conductance drift, read noise, device variation) do not degrade performance; no cross-domain or hardware-in-the-loop experiments are described.
minor comments (1)
  1. [Abstract] Abstract: the spelling 'frontalisation' appears; ensure consistent British/American spelling throughout the manuscript.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. We address the two major comments below. Where the manuscript is deficient in supporting details or validation, we will revise accordingly rather than defend unsubstantiated claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim of 'up to 96% identification accuracy' is asserted without any description of the two datasets, baselines, training protocols, error bars, or hardware implementation, supplying no evidence that the claim is supported.

    Authors: We agree the abstract is overly concise and does not supply the requested context. The full manuscript contains sections describing the two datasets, the GAN and memristor training protocols, and the accuracy results. To address the concern directly, we will revise the abstract to name the datasets, indicate that results include comparison against baselines, and state that detailed protocols and any error statistics appear in the experimental section. Hardware implementation is performed via device-level simulation of the memristor model rather than physical deployment; this will be clarified in both abstract and methods. revision: yes

  2. Referee: [Proposed framework] Proposed framework (integration section): the claim that memristor-based classifiers retain high accuracy on GAN-synthesized frontal faces rests on the untested assumptions that (1) GAN outputs lie in the same distribution as real frontal training data and (2) memristor non-idealities (conductance drift, read noise, device variation) do not degrade performance; no cross-domain or hardware-in-the-loop experiments are described.

    Authors: The comment correctly identifies that the manuscript does not present explicit cross-domain distribution tests or hardware-in-the-loop measurements of non-idealities. The reported 96% figure is obtained from end-to-end simulation in which the memristor classifier is evaluated on GAN outputs, but without separate verification that the synthetic images match the real frontal distribution or that device variation has been injected. We will revise the integration section to state these assumptions explicitly, add a limitations paragraph, and either include new distribution-comparison metrics (if they can be computed from existing data) or qualify the accuracy claim as simulation-based only. Hardware-in-the-loop validation is outside the current scope and will be noted as future work. revision: partial

Circularity Check

0 steps flagged

No circularity; absence of any derivation chain or equations

full rationale

The manuscript is a descriptive systems paper proposing an integration of GAN-based frontalization with memristor classifiers and reporting empirical accuracies up to 96%. It contains no equations, no claimed first-principles derivations, and no load-bearing mathematical steps that could reduce to their own inputs. Consequently no circularity patterns (self-definitional, fitted-input-as-prediction, or self-citation chains) are present, and the work is self-contained as an engineering demonstration rather than a deductive argument.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The work is an applied engineering demonstration with no explicit mathematical axioms, free parameters, or invented physical entities described in the abstract.

pith-pipeline@v0.9.1-grok · 5697 in / 1001 out tokens · 19327 ms · 2026-06-27T10:10:53.881975+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

101 extracted references · 37 canonical work pages

  1. [1]

    , Matthews , I

    bchapter Gross , R. , Matthews , I. , Cohn , J. , Kanade , T. , Baker , S. : Multi-pie . In: 2008 8th IEEE International Conference on Automatic Face & Gesture Recognition , pp. 1 -- 8 ( 2008 ). 10.1109/AFGR.2008.4813399 bchapter

    work page doi:10.1109/afgr.2008.4813399 2008

  2. [3]

    , Huang , J.-B

    botherref Lai , W.-S. , Huang , J.-B. , Ahuja , N. , Yang , M.-H. : Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution (2017). https://arxiv.org/abs/1704.03915 botherref

    Pith/arXiv arXiv 2017

  3. [4]

    , Serb , A

    botherref Messaris , I. , Serb , A. , Khiat , A. , Nikolaidis , S. , Prodromakis , T. : A compact Verilog-A ReRAM switching model (2017). https://arxiv.org/abs/1703.01167 botherref

    Pith/arXiv arXiv 2017

  4. [5]

    , Guo , J

    bchapter Deng , J. , Guo , J. , Xue , N. , Zafeiriou , S. : Arcface: Additive angular margin loss for deep face recognition . In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) ( 2019 ) bchapter

    2019

  5. [6]

    , Zisserman , A

    botherref Simonyan , K. , Zisserman , A. : Very Deep Convolutional Networks for Large-Scale Image Recognition (2015). https://arxiv.org/abs/1409.1556 botherref

    Pith/arXiv arXiv 2015

  6. [7]

    Multi-PIE , year=

    Gross, Ralph and Matthews, Iain and Cohn, Jeffrey and Kanade, Takeo and Baker, Simon , booktitle=. Multi-PIE , year=

  7. [8]

    Proceedings of the 8th ACM on Multimedia Systems Conference , pages =

    Hsu, Hwai-Jung and Chen, Kuan-Ta , title =. Proceedings of the 8th ACM on Multimedia Systems Conference , pages =. 2017 , isbn =. doi:10.1145/3083187.3083214 , abstract =

    work page doi:10.1145/3083187.3083214 2017

  8. [9]

    2017 , eprint=

    Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution , author=. 2017 , eprint=

    2017

  9. [10]

    2017 , eprint=

    A compact Verilog-A ReRAM switching model , author=. 2017 , eprint=

    2017

  10. [11]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Deng, Jiankang and Guo, Jia and Xue, Niannan and Zafeiriou, Stefanos , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

  11. [12]

    2015 , eprint=

    Very Deep Convolutional Networks for Large-Scale Image Recognition , author=. 2015 , eprint=

    2015

  12. [15]

    UAV surveillance for violence detection and individual identification , volume =

    Srivastava, Anugrah and Badal, Tapas and Saxena, Pawan and Vidyarthi, Ankit and Singh, Rishav , year =. UAV surveillance for violence detection and individual identification , volume =. Automated Software Engineering , doi =

  13. [16]

    Forensic face recognition as a means to determine strength of evidence: A survey , volume =

    Zeinstra, C and Meuwly, Didier and Ruifrok, Arnout and Veldhuis, Raymond and Spreeuwers, Luuk , year =. Forensic face recognition as a means to determine strength of evidence: A survey , volume =

  14. [17]

    Facial recognition application for border control , year=

    Carlos-Roca, Laura Rodríguez and Torres, Isabelle Hupont and Tena, Carles Fernández , booktitle=. Facial recognition application for border control , year=

  15. [19]

    2024 , publisher =

    Lowell, Theodore Sorenson , title =. 2024 , publisher =

    2024

  16. [20]

    Machine Learning in Computer Vision , journal =

    Asharul Islam Khan and Salim Al-Habsi , keywords =. Machine Learning in Computer Vision , journal =. 2020 , note =. doi:https://doi.org/10.1016/j.procs.2020.03.355 , url =

    work page doi:10.1016/j.procs.2020.03.355 2020

  17. [21]

    and Khemani, Bharti and Patil, Shruti and Kotecha, Ketan and Vora, Deepali R

    Mahadevkar, Supriya V. and Khemani, Bharti and Patil, Shruti and Kotecha, Ketan and Vora, Deepali R. and Abraham, Ajith and Gabralla, Lubna Abdelkareim , journal=. A Review on Machine Learning Styles in Computer Vision—Techniques and Future Directions , year=

  18. [22]

    Baseer: A Drone-based System Leveraging Facial and Pattern Recognition for Efficient Real-Time Child Recovery , volume =

    Attiah, Afraa and Hakeem, Abeer , year =. Baseer: A Drone-based System Leveraging Facial and Pattern Recognition for Efficient Real-Time Child Recovery , volume =. Journal of King Abdulaziz University: Computing and Information Technology Sciences , doi =

  19. [23]

    2023 , eprint=

    DroneNet: Crowd Density Estimation using Self-ONNs for Drones , author=. 2023 , eprint=

    2023

  20. [24]

    Surveillance and violence from afar: The politics of drones and liminal security-scapes , volume =

    Wall, Tyler and Monahan, Torin , year =. Surveillance and violence from afar: The politics of drones and liminal security-scapes , volume =. Theoretical Criminology , doi =

  21. [25]

    Security analysis of drones systems: Attacks, limitations, and recommendations , journal =

    Jean-Paul Yaacoub and Hassan Noura and Ola Salman and Ali Chehab , keywords =. Security analysis of drones systems: Attacks, limitations, and recommendations , journal =. 2020 , issn =. doi:https://doi.org/10.1016/j.iot.2020.100218 , url =

    work page doi:10.1016/j.iot.2020.100218 2020

  22. [26]

    Security Dialogue , volume =

    Katharine Hall Kindervater , title =. Security Dialogue , volume =. 2016 , doi =. https://doi.org/10.1177/0967010615616011 , abstract =

    work page doi:10.1177/0967010615616011 2016

  23. [27]

    Proceedings of the First Workshop on Micro Aerial Vehicle Networks, Systems, and Applications for Civilian Use , pages =

    Hsu, Hwai-Jung and Chen, Kuan-Ta , title =. Proceedings of the First Workshop on Micro Aerial Vehicle Networks, Systems, and Applications for Civilian Use , pages =. 2015 , isbn =. doi:10.1145/2750675.2750679 , abstract =

    work page doi:10.1145/2750675.2750679 2015

  24. [28]

    Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis , year=

    Huang, Rui and Zhang, Shu and Li, Tianyu and He, Ran , booktitle=. Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis , year=

  25. [29]

    Face recognition using deep multi-pose representations , year=

    AbdAlmageed, Wael and Wu, Yue and Rawls, Stephen and Harel, Shai and Hassner, Tal and Masi, Iacopo and Choi, Jongmoo and Lekust, Jatuporn and Kim, Jungyeon and Natarajan, Prem and Nevatia, Ram and Medioni, Gerard , booktitle=. Face recognition using deep multi-pose representations , year=

  26. [30]

    , journal=

    Zhang, Zhihong and Chen, Xu and Wang, Beizhan and Hu, Guosheng and Zuo, Wangmeng and Hancock, Edwin R. , journal=. Face Frontalization Using an Appearance-Flow-Based Convolutional Neural Network , year=

  27. [31]

    Procedia Computer Science , year=

    AirNav:A Computation-driven Cloud-based Solution for the Drone Path Planning Utilizing Machine Learning , author=. Procedia Computer Science , year=

  28. [32]

    UAV-Based IoT Platform: A Crowd Surveillance Use Case , year=

    Motlagh, Naser Hossein and Bagaa, Miloud and Taleb, Tarik , journal=. UAV-Based IoT Platform: A Crowd Surveillance Use Case , year=

  29. [34]

    2022 , eprint=

    The Computational Limits of Deep Learning , author=. 2022 , eprint=

    2022

  30. [36]

    Micromachines , VOLUME =

    Chai, Qiuyue and Liu, Yue , TITLE =. Micromachines , VOLUME =. 2024 , NUMBER =

    2024

  31. [37]

    A Compact Memristor-based GAN Architecture with a Case Study on Single Image Super-Resolution , year=

    Dong, Zhekang and Fang, Yixiao and Huang, Ling and Li, Jidong and Qi, Donglian , booktitle=. A Compact Memristor-based GAN Architecture with a Case Study on Single Image Super-Resolution , year=

  32. [38]

    2019 , eprint=

    A Memristor based Unsupervised Neuromorphic System Towards Fast and Energy-Efficient GAN , author=. 2019 , eprint=

    2019

  33. [39]

    2016 , eprint=

    Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , author=. 2016 , eprint=

    2016

  34. [41]

    SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation , year=

    Badrinarayanan, Vijay and Kendall, Alex and Cipolla, Roberto , journal=. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation , year=

  35. [42]

    2019 , eprint=

    PyTorch: An Imperative Style, High-Performance Deep Learning Library , author=. 2019 , eprint=

    2019

  36. [43]

    On Depth and Complexity of Generative Adversarial Networks , keywords =

    Yamazaki, Hiroyuki Vincent , institution =. On Depth and Complexity of Generative Adversarial Networks , keywords =. 2017 , url =

    2017

  37. [44]

    A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects , year=

    Li, Zewen and Liu, Fan and Yang, Wenjie and Peng, Shouheng and Zhou, Jun , journal=. A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects , year=

  38. [45]

    Rectifier Nonlinearities Improve Neural Network Acoustic Models , author=. Proc. icml , volume=. 2013 , url=

    2013

  39. [46]

    , booktitle=

    Isola, Phillip and Zhu, Jun-Yan and Zhou, Tinghui and Efros, Alexei A. , booktitle=. Image-to-Image Translation with Conditional Adversarial Networks , year=

  40. [48]

    PyT-NeuroPack: A Hybrid PyTorch/Memristor-Crossbar Simulation Tool for Convolutional Neural Networks , year=

    Sestito, Cristian and Huang, Weijie and Agwa, Shady and Prodromakis, Themis , booktitle=. PyT-NeuroPack: A Hybrid PyTorch/Memristor-Crossbar Simulation Tool for Convolutional Neural Networks , year=

  41. [49]

    A Threshold Selection Method from Gray-Level Histograms , year=

    Otsu, Nobuyuki , journal=. A Threshold Selection Method from Gray-Level Histograms , year=

  42. [50]

    Adrian, E. D. and Zotterman, Yngve , title =. The Journal of Physiology , volume =. doi:https://doi.org/10.1113/jphysiol.1926.sp002308 , url =. https://physoc.onlinelibrary.wiley.com/doi/pdf/10.1113/jphysiol.1926.sp002308 , year =

    work page doi:10.1113/jphysiol.1926.sp002308 1926

  43. [51]

    Excitatory and Inhibitory Memristive Synapses for Spiking Neural Networks , year=

    Lecerf, Gwendal and Tomas, Jean and Saïghi, Sylvain , booktitle=. Excitatory and Inhibitory Memristive Synapses for Spiking Neural Networks , year=

  44. [52]

    Homogeneous Synaptic Input , author=

    A Review of the Integrate-and-fire Neuron Model: I. Homogeneous Synaptic Input , author=. Biological Cybernetics , year=

  45. [53]

    2006 , Journal =

    The tempotron: a neuron that learns spike timing-based decisions , Author =. 2006 , Journal =. doi:10.1038/nn1643 , Number =

    work page doi:10.1038/nn1643 2006

  46. [54]

    , journal=

    Izhikevich, E.M. , journal=. Simple model of spiking neurons , year=

  47. [55]

    Frontiers in Neuroscience , year=

    Learning Without Feedback: Fixed Random Learning Signals Allow for Feedforward Training of Deep Neural Networks , author=. Frontiers in Neuroscience , year=

  48. [56]

    PLOS Computational Biology , publisher =

    Unsupervised Learning of Visual Features through Spike Timing Dependent Plasticity , year =. PLOS Computational Biology , publisher =. doi:10.1371/journal.pcbi.0030031 , author =

    work page doi:10.1371/journal.pcbi.0030031

  49. [57]

    Frontiers in Neuroscience , year=

    Enabling Spike-Based Backpropagation for Training Deep Neural Network Architectures , author=. Frontiers in Neuroscience , year=

  50. [58]

    2018 , eprint=

    Activation Functions: Comparison of trends in Practice and Research for Deep Learning , author=. 2018 , eprint=

    2018

  51. [59]

    , TITLE =

    Abidi, Syed Murtaza Hussain and Hassan, Syed Ali and Raza, Syed Muhammad and Beliatis, Michail J. , TITLE =. Electronics , VOLUME =. 2026 , NUMBER =

    2026

  52. [60]

    Memristor-based hardware accelerators for artificial intelligence , volume =

    Huang, Yi and Ando, Takashi and Sebastian, Abu and Chang, Meng-Fan and Yang, Jianhua Joshua and Xia, Qiangfei , year =. Memristor-based hardware accelerators for artificial intelligence , volume =. Nature Reviews Electrical Engineering , doi =

  53. [61]

    Nanomaterials , VOLUME =

    Wang, Xiangjing and Zhu, Yixin and Zhou, Zili and Chen, Xin and Jia, Xiaojun , TITLE =. Nanomaterials , VOLUME =. 2025 , NUMBER =

    2025

  54. [62]

    Aguirre, Fernando and Sebastian, Abu and Gallo, Manuel and Song, Wenhao and Wang, Tong and Yang, Jianhua Joshua and Lu, Wei and Chang, Meng-Fan and Ielmini, D. and Yang, Yuchao and Mehonic, Adnan and Kenyon, Anthony and Villena, Marco and Roldan, Juan and Wu, Yuting and Hsu, Hung-Hsi and Raghavan, Nagarajan and Sune, Jordi and Miranda, Enrique and Lanza, ...

  55. [63]

    A Survey on the Convergence of Edge Computing and AI for UAVs: Opportunities and Challenges , year=

    McEnroe, Patrick and Wang, Shen and Liyanage, Madhusanka , journal=. A Survey on the Convergence of Edge Computing and AI for UAVs: Opportunities and Challenges , year=

  56. [64]

    Applied Sciences , VOLUME =

    Zhalgas, Aidana and Amirgaliyev, Beibut and Sovet, Adil , TITLE =. Applied Sciences , VOLUME =. 2025 , NUMBER =

    2025

  57. [65]

    2018 , eprint=

    Surveillance Face Recognition Challenge , author=. 2018 , eprint=

    2018

  58. [66]

    Unconstrained Face Identification using Ensembles trained on Clustered Data , year=

    Vareto, Rafael Henrique and Schwartz, William Robson , booktitle=. Unconstrained Face Identification using Ensembles trained on Clustered Data , year=

  59. [67]

    2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) , year=

    Toward Open-Set Face Recognition , author=. 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) , year=

    2017

  60. [68]

    Recent Advances in Open Set Recognition: A Survey , year=

    Geng, Chuanxing and Huang, Sheng-Jun and Chen, Songcan , journal=. Recent Advances in Open Set Recognition: A Survey , year=

  61. [69]

    Empowering generative AI through mobile edge computing , volume =

    Ale, Laha and Zhang, Ning and King, Scott and Chen, Dajiang , year =. Empowering generative AI through mobile edge computing , volume =. Nature Reviews Electrical Engineering , doi =

  62. [70]

    and Gopi, R

    Suganya, B. and Gopi, R. and Anandan, Ranjith and Singh, Gavendra , year =. Dynamic task offloading edge-aware optimization framework for enhanced UAV operations on edge computing platform , volume =. Scientific Reports , doi =

  63. [72]

    A -Controller-Based System for Interfacing Selectorless RRAM Crossbar Arrays , year=

    Berdan, Radu and Serb, Alexander and Khiat, Ali and Regoutz, Anna and Papavassiliou, Christos and Prodromakis, Themis , journal=. A -Controller-Based System for Interfacing Selectorless RRAM Crossbar Arrays , year=

  64. [74]

    , Zhan , X

    barticle Wang , Y. , Zhan , X. , Zhan , T. , Xu , J. , Bai , X. : Machine learning-based facial recognition for financial fraud prevention . Journal of Computer Technology and Applied Mathematics 1 ( 1 ), 77 -- 84 ( 2024 ) 10.5281/zenodo.11004115 barticle

    work page doi:10.5281/zenodo.11004115 2024

  65. [75]

    , Naveen Kumar , S

    barticle Rameswari , R. , Naveen Kumar , S. , Abishek Aananth , M. , Deepak , C. : Automated access control system using face recognition . Materials Today: Proceedings 45 , 1251 -- 1256 ( 2021 ) 10.1016/j.matpr.2020.04.664 . International Conference on Advances in Materials Research - 2019 barticle

    work page doi:10.1016/j.matpr.2020.04.664 2021

  66. [76]

    , Badal , T

    botherref Srivastava , A. , Badal , T. , Saxena , P. , Vidyarthi , A. , Singh , R. : Uav surveillance for violence detection and individual identification. Automated Software Engineering 29 (2022) 10.1007/s10515-022-00323-3 botherref

    work page doi:10.1007/s10515-022-00323-3 2022

  67. [77]

    , Meuwly , D

    barticle Zeinstra , C. , Meuwly , D. , Ruifrok , A. , Veldhuis , R. , Spreeuwers , L. : Forensic face recognition as a means to determine strength of evidence: A survey . Forensic science review 30 , 21 -- 32 ( 2018 ) barticle

    2018

  68. [78]

    , Torres , I.H

    bchapter Carlos-Roca , L.R. , Torres , I.H. , Tena , C.F. : Facial recognition application for border control . In: 2018 International Joint Conference on Neural Networks (IJCNN) , pp. 1 -- 7 ( 2018 ). 10.1109/IJCNN.2018.8489113 bchapter

    work page doi:10.1109/ijcnn.2018.8489113 2018

  69. [79]

    : Facial Biometric Authentication for Smartphones: The Intersection of Security and Usability

    botherref Lowell , T.S. : Facial Biometric Authentication for Smartphones: The Intersection of Security and Usability. The University of Arizona (2024). http://hdl.handle.net/10150/672890 botherref

    2024

  70. [80]

    , Moctezuma , D

    barticle Sanchez del Rio , J. , Moctezuma , D. , Conde , C. , Martin de Diego , I. , Cabello , E. : Automated border control e-gates and facial recognition systems . Computers & Security 62 , 49 -- 72 ( 2016 ) 10.1016/j.cose.2016.07.001 barticle

    work page doi:10.1016/j.cose.2016.07.001 2016

  71. [81]

    , Amirgaliyev , B

    botherref Zhalgas , A. , Amirgaliyev , B. , Sovet , A. : Robust face recognition under challenging conditions: A comprehensive review of deep learning methods and challenges. Applied Sciences 15(17) (2025) 10.3390/app15179390 botherref

    work page doi:10.3390/app15179390 2025

  72. [82]

    , Zhu , X

    botherref Cheng , Z. , Zhu , X. , Gong , S. : Surveillance Face Recognition Challenge (2018). https://arxiv.org/abs/1804.09691 botherref

    Pith/arXiv arXiv 2018

  73. [83]

    , Zhang , S

    bchapter Huang , R. , Zhang , S. , Li , T. , He , R. : Beyond face rotation: Global and local perception gan for photorealistic and identity preserving frontal view synthesis . In: 2017 IEEE International Conference on Computer Vision (ICCV) , pp. 2458 -- 2467 ( 2017 ). 10.1109/ICCV.2017.267 bchapter

    work page doi:10.1109/iccv.2017.267 2017

  74. [84]

    , Wu , Y

    bchapter AbdAlmageed , W. , Wu , Y. , Rawls , S. , Harel , S. , Hassner , T. , Masi , I. , Choi , J. , Lekust , J. , Kim , J. , Natarajan , P. , Nevatia , R. , Medioni , G. : Face recognition using deep multi-pose representations . In: 2016 IEEE Winter Conference on Applications of Computer Vision (WACV) , pp. 1 -- 9 ( 2016 ). 10.1109/WACV.2016.7477555 bchapter

    work page doi:10.1109/wacv.2016.7477555 2016

  75. [85]

    , Chen , X

    barticle Zhang , Z. , Chen , X. , Wang , B. , Hu , G. , Zuo , W. , Hancock , E.R. : Face frontalization using an appearance-flow-based convolutional neural network . IEEE Transactions on Image Processing 28 ( 5 ), 2187 -- 2199 ( 2019 ) 10.1109/TIP.2018.2883554 barticle

    work page doi:10.1109/tip.2018.2883554 2019

  76. [86]

    , Wang , S

    barticle McEnroe , P. , Wang , S. , Liyanage , M. : A survey on the convergence of edge computing and ai for uavs: Opportunities and challenges . IEEE Internet of Things Journal 9 ( 17 ), 15435 -- 15459 ( 2022 ) 10.1109/JIOT.2022.3176400 barticle

    work page doi:10.1109/jiot.2022.3176400 2022

  77. [87]

    , Gopi , R

    botherref Suganya , B. , Gopi , R. , Anandan , R. , Singh , G. : Dynamic task offloading edge-aware optimization framework for enhanced uav operations on edge computing platform. Scientific Reports 14 (2024) 10.1038/s41598-024-67285-2 botherref

    work page doi:10.1038/s41598-024-67285-2 2024

  78. [88]

    , Bhavsar , M

    botherref Dadheech , A. , Bhavsar , M. : Airnav:a computation-driven cloud-based solution for the drone path planning utilizing machine learning. Procedia Computer Science (2025) botherref

    2025

  79. [89]

    , Bagaa , M

    barticle Motlagh , N.H. , Bagaa , M. , Taleb , T. : Uav-based iot platform: A crowd surveillance use case . IEEE Communications Magazine 55 ( 2 ), 128 -- 134 ( 2017 ) 10.1109/MCOM.2017.1600587CM barticle

    work page doi:10.1109/mcom.2017.1600587cm 2017

  80. [90]

    , Truong , L.H

    barticle Do , H.T. , Truong , L.H. , Nguyen , M.T. , Chien , C.-F. , Tran , H.T. , Hua , H.T. , Nguyen , C.V. , Nguyen , H.T.T. , Nguyen , N.T.T. : Energy-efficient unmanned aerial vehicle (uav) surveillance utilizing artificial intelligence (ai) . Wireless Communications and Mobile Computing 2021 ( 1 ), 8615367 ( 2021 ) 10.1155/2021/8615367 https://arxiv...

    work page doi:10.1155/2021/8615367 2021

Showing first 80 references.