Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
173 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Unified Negative Pair Generation toward Well-discriminative Feature Space for Face Recognition (2203.11593v2)

Published 22 Mar 2022 in cs.CV

Abstract: The goal of face recognition (FR) can be viewed as a pair similarity optimization problem, maximizing a similarity set $\mathcal{S}p$ over positive pairs, while minimizing similarity set $\mathcal{S}n$ over negative pairs. Ideally, it is expected that FR models form a well-discriminative feature space (WDFS) that satisfies $\inf{\mathcal{S}p} > \sup{\mathcal{S}n}$. With regard to WDFS, the existing deep feature learning paradigms (i.e., metric and classification losses) can be expressed as a unified perspective on different pair generation (PG) strategies. Unfortunately, in the metric loss (ML), it is infeasible to generate negative pairs taking all classes into account in each iteration because of the limited mini-batch size. In contrast, in classification loss (CL), it is difficult to generate extremely hard negative pairs owing to the convergence of the class weight vectors to their center. This leads to a mismatch between the two similarity distributions of the sampled pairs and all negative pairs. Thus, this paper proposes a unified negative pair generation (UNPG) by combining two PG strategies (i.e., MLPG and CLPG) from a unified perspective to alleviate the mismatch. UNPG introduces useful information about negative pairs using MLPG to overcome the CLPG deficiency. Moreover, it includes filtering the similarities of noisy negative pairs to guarantee reliable convergence and improved performance. Exhaustive experiments show the superiority of UNPG by achieving state-of-the-art performance across recent loss functions on public benchmark datasets. Our code and pretrained models are publicly available.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. Vggface2: A dataset for recognising faces across pose and age. In 2018 13th IEEE international conference on automatic face & gesture recognition (FG 2018), pages 67–74. IEEE, 2018.
  2. K-face: A large-scale kist face database in consideration with unconstrained environments. arXiv preprint arXiv:2103.02211, 2021.
  3. Learning a similarity metric discriminatively, with application to face verification. In CVPR, volume 1, pages 539–546. IEEE, 2005.
  4. Arcface: Additive angular margin loss for deep face recognition. In CVPR, pages 4690–4699, 2019.
  5. Ms-celeb-1m: A dataset and benchmark for large-scale face recognition. In ECCV, pages 87–102. Springer, 2016.
  6. Dimensionality reduction by learning an invariant mapping. In CVPR, volume 2, pages 1735–1742. IEEE, 2006.
  7. Deep residual learning for image recognition. In CVPR, pages 770–778, 2016.
  8. E. Hoffer and N. Ailon. Deep metric learning using triplet network. In International workshop on similarity-based pattern recognition, pages 84–92. Springer, 2015.
  9. Squeeze-and-excitation networks. In CVPR, pages 7132–7141, 2018.
  10. Labeled faces in the wild: A database forstudying face recognition in unconstrained environments. In Workshop on faces in’Real-Life’Images: detection, alignment, and recognition, 2008.
  11. Mixface: Improving face verification focusing on fine-grained conditions. arXiv preprint arXiv:2111.01717, 2021.
  12. The megaface benchmark: 1 million faces for recognition at scale. In CVPR, pages 4873–4882, 2016.
  13. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25:1097–1105, 2012.
  14. Sphereface: Deep hypersphere embedding for face recognition. In CVPR, pages 212–220, 2017.
  15. Large-margin softmax loss for convolutional neural networks. In ICML, volume 2, page 7, 2016.
  16. I. Loshchilov and F. Hutter. Sgdr: Stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983, 2016.
  17. Iarpa janus benchmark-c: Face dataset and protocol. In 2018 International Conference on Biometrics, pages 158–165. IEEE, 2018.
  18. Magface: A universal representation for face recognition and quality assessment. In CVPR, pages 14225–14234, 2021.
  19. Agedb: the first manually collected, in-the-wild age database. In CVPRW, pages 51–59, 2017.
  20. Deep metric learning via lifted structured feature embedding. In CVPR, pages 4004–4012, 2016.
  21. Deep face recognition. In BMVC, pages 41.1–41.12, 2015.
  22. L2-constrained softmax loss for discriminative face verification. arXiv preprint arXiv:1703.09507, 2017.
  23. Facenet: A unified embedding for face recognition and clustering. In CVPR, pages 815–823, 2015.
  24. Frontal to profile face verification in the wild. In 2016 IEEE Winter Conference on Applications of Computer Vision, pages 1–9. IEEE, 2016.
  25. K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.
  26. K. Sohn. Improved deep metric learning with multi-class n-pair loss objective. In NeurIPS, pages 1857–1865, 2016.
  27. Circle loss: A unified perspective of pair similarity optimization. In CVPR, pages 6398–6407, 2020.
  28. Deeply learned face representations are sparse, selective, and robust. In CVPR, pages 2892–2900, 2015.
  29. Going deeper with convolutions. In CVPR, pages 1–9, 2015.
  30. Deepface: Closing the gap to human-level performance in face verification. In CVPR, pages 1701–1708, 2014.
  31. J. W. Tukey et al. Exploratory data analysis, volume 2. Reading, Mass., 1977.
  32. Additive margin softmax for face verification. IEEE Sign. Process. Letters, 25(7):926–930, 2018.
  33. Normface: L2 hypersphere embedding for face verification. In ACM MM, pages 1041–1049, 2017.
  34. Cosface: Large margin cosine loss for deep face recognition. In CVPR, pages 5265–5274, 2018.
  35. Multi-similarity loss with general pair weighting for deep metric learning. In CVPR, pages 5022–5030, 2019.
  36. A discriminative feature learning approach for deep face recognition. In ECCV, pages 499–515. Springer, 2016.
  37. Iarpa janus benchmark-b face dataset. In CVPRW, pages 90–98, 2017.
  38. Learning face representation from scratch. arXiv preprint arXiv:1411.7923, 2014.
  39. B. Yu and D. Tao. Deep metric learning with tuplet margin loss. In CVPR, pages 6490–6499, 2019.
  40. Adacos: Adaptively scaling cosine logits for effectively learning deep face representations. In CVPR, pages 10823–10832, 2019.
  41. Cross-age lfw: A database for studying cross-age face recognition in unconstrained environments. arXiv preprint arXiv:1708.08197, 2017.
Citations (4)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets