Deepface: Closing the gap to human-level performance in face verification. In modern face recognition, the conventional pipeline consists of four stages: detect => align => represent => classify. We revisit both the alignment step and the representation step by employing explicit 3D face modeling in order to apply a piecewise affine transformation, and derive a face representation from a nine-layer deep neural network. This deep network involves more than 120 million parameters using several locally connected layers without weight sharing, rather than the standard convolutional layers. Thus we trained it on the largest facial dataset to-date, an identity labeled dataset of four million facial images belonging to more than 4, 000 identities. The learned representations coupling the accurate model-based alignment with the large facial database generalize remarkably well to faces in unconstrained environments, even with a simple classifier. Our method reaches an accuracy of 97.35% on the Labeled Faces in the Wild (LFW) dataset, reducing the error of the current state of the art by more than 27%, closely approaching human-level performance.

References in zbMATH (referenced in 20 articles )

Showing results 1 to 20 of 20.
Sorted by year (citations)

  1. Wäldchen, Stephan; Macdonald, Jan; Hauch, Sascha; Kutyniok, Gitta: The computational complexity of understanding binary classifier decisions (2021)
  2. Duan, Jia; Zhou, Jiantao; Li, Yuanman: Privacy-preserving distributed deep learning based on secret sharing (2020)
  3. Liu, Minliang; Liang, Liang; Sun, Wei: A generic physics-informed neural network-based constitutive model for soft biological tissues (2020)
  4. Pai, Gautam; Joseph-Rivlin, Mor; Kimmel, Ron; Sochen, Nir: On geometric invariants, learning, and recognition of shapes and forms (2020)
  5. Sirignano, Justin; Spiliopoulos, Konstantinos: Mean field analysis of neural networks: a law of large numbers (2020)
  6. Ghatak, Abhijit: Deep learning with R (2019)
  7. Görgel, Pelin; Simsek, Ahmet: Face recognition via deep stacked denoising sparse autoencoders (DSDSA) (2019)
  8. Liu, Minliang; Liang, Liang; Sun, Wei: Estimation of in vivo constitutive parameters of the aortic wall using a machine learning approach (2019)
  9. Viroli, Cinzia; McLachlan, Geoffrey J.: Deep Gaussian mixture models (2019)
  10. Wang, Shanshan; Chen, Ying: A joint loss function for deep face recognition (2019)
  11. Xue, Shan; Zhu, Hong: Low-resolution and open-set face recognition via recursive label propagation based on statistical classification (2019)
  12. Savvopoulos, Alkiviadis; Kanavos, Andreas; Mylonas, Phivos; Sioutas, Spyros: LSTM accelerator for convolutional object identification (2018)
  13. Zhu, Xudong; Li, Hui; Yu, Yang: Blockchain-based privacy preserving deep learning (2018)
  14. Rawat, Waseem; Wang, Zenghui: Deep convolutional neural networks for image classification: a comprehensive review (2017)
  15. Savchenko, Andrey V.: Clustering and maximum likelihood search for efficient statistical classification with medium-sized databases (2017)
  16. Szczodrak, Maciej; Czyżewski, Andrzej: Evaluation of face detection algorithms for the bank client identity verification (2017)
  17. Cheng, Xiuyuan; Chen, Xu; Mallat, Stéphane: Deep Haar scattering networks (2016)
  18. Galanti, Tomer; Wolf, Lior; Hazan, Tamir: A theoretical framework for deep transfer learning (2016)
  19. Xin Liu, Meina Kan, Wanglong Wu, Shiguang Shan, Xilin Chen: VIPLFaceNet: An Open Source Deep Face Recognition SDK (2016) arXiv
  20. Wang, Guoyin; Xu, Ji; Zhang, Qinghua; Liu, Yuchao: Multi-granularity intelligent information processing (2015)