Skip to main content
Book cover

European Conference on Computer Vision

ECCV 2020: Computer Vision – ECCV 2020 Workshops pp 463–481Cite as

FairFace Challenge at ECCV 2020: Analyzing Bias in Face Recognition

Part of the Lecture Notes in Computer Science book series (LNIP,volume 12540)

Abstract

This work summarizes the 2020 ChaLearn Looking at People Fair Face Recognition and Analysis Challenge and provides a description of the top-winning solutions and analysis of the results. The aim of the challenge was to evaluate accuracy and bias in gender and skin colour of submitted algorithms on the task of 1:1 face verification in the presence of other confounding attributes. Participants were evaluated using an in-the-wild dataset based on reannotated IJB-C, further enriched 12.5K new images and additional labels. The dataset is not balanced, which simulates a real world scenario where AI-based models supposed to present fair outcomes are trained and evaluated on imbalanced data. The challenge attracted 151 participants, who made more 1.8K submissions in total. The final phase of the challenge attracted 36 active teams out of which 10 exceeded 0.999 AUC-ROC while achieving very low scores in the proposed bias metrics. Common strategies by the participants were face pre-processing, homogenization of data distributions, the use of bias aware loss functions and ensemble models. The analysis of top-10 teams shows higher false positive rates (and lower false negative rates) for females with dark skin tone as well as the potential of eyeglasses and young age to increase the false positive rates too.

Keywords

  • Face verification
  • Face recognition
  • Fairness
  • Bias

These (T. Sixta and J. C. S. Jacques Junior) authors contributed equally to this work.

This is a preview of subscription content, access via your institution.

Chapter
USD   29.95
Price excludes VAT (USA)
  • DOI: 10.1007/978-3-030-65414-6_32
  • Chapter length: 19 pages
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
eBook
USD   89.00
Price excludes VAT (USA)
  • ISBN: 978-3-030-65414-6
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
Softcover Book
USD   119.99
Price excludes VAT (USA)
Learn about institutional subscriptions
Fig. 1.
Fig. 2.
Fig. 3.
Fig. 4.

Notes

  1. 1.

    Data and winning solutions codes are available at http://chalearnlap.cvc.uab.es/challenge/38/description.

  2. 2.

    For more information about ethics in AI you can visit the European guideline in the following link https://ec.europa.eu/digital-single-market/en/news/ethics-guidelines-trustworthy-ai.

  3. 3.

    https://competitions.codalab.org/competitions/24184.

  4. 4.

    https://competitions.codalab.org.

  5. 5.

    Attribute categories used in this work are imperfect for many reasons. For example, it is unclear how many skin colour and gender categories should be stipulated (or whether they should be treated as discrete categories at all). We base our definitions on widely accepted traditional categories and our methodology and findings are expected to be applied later to any re-defined and/or extended attribute category.

  6. 6.

    The full leaderboards for both phases are shown in the supplementary material.

  7. 7.

    https://github.com/paranoidai/Fairface-Recognition-Solution.

  8. 8.

    https://github.com/HaoSir/ECCV-2020-Fair-Face-Recognition-challenge_2nd_place_solution-ustc-nelslip-.

  9. 9.

    https://github.com/CdtQin/FairFace.

References

  1. Facial recognition tech under spotlight after Boston bombings. Biometric Technology Today 2013(5), 1 (2013)

    Google Scholar 

  2. Albiero, V., Bowyer, K.W., Vangara, K., King, M.C.: Does face recognition accuracy get better with age? Deep face matchers say no. In: Winter Conference on Applications of Computer Vision (WACV), pp. 250–258 (2020)

    Google Scholar 

  3. Albiero, V., Krishnapriya, K.S., Vangara, K., Zhang, K., King, M.C., Bowyer, K.W.: Analysis of gender inequality in face recognition accuracy. CoRR abs/2002.00065 (2020)

    Google Scholar 

  4. Alvi, M.S., Zisserman, A., Nellåker, C.: Turning a blind eye: explicit removal of biases and variation from deep neural network embeddings. CoRR abs/1809.02169 (2018)

    Google Scholar 

  5. Snow, J.: Amazon’s Face Recognition Falsely Matched 28 Members of Congress With Mugshots. American Civil Liberties Union, July 2018. https://www.aclu.org/blog/privacy-technology/surveillance-technologies/amazons-face-recognition-falsely-matched-28. Accessed 5 Aug 2020

  6. Anne Hendricks, L., Burns, K., Saenko, K., Darrell, T., Rohrbach, A.: Women also snowboard: Overcoming bias in captioning models. In: European Conference on Computer Vision (ECCV), pp. 793–811 (2018)

    Google Scholar 

  7. Bellamy, R.K.E., et al.: AI fairness 360: an extensible toolkit for detecting, understanding, and mitigating unwanted algorithmic bias. CoRR abs/1810.01943 (2018)

    Google Scholar 

  8. Berk, R., Heidari, H., Jabbari, S., Kearns, M., Roth, A.: Fairness in criminal justice risk assessments: the state of the art. Sociological Methods & Research (2018)

    Google Scholar 

  9. Bino, S., Bernerd, F.: Variations in skin colour and the biological consequences of ultraviolet radiation exposure. Br. J. Dermatol. 169(s3), 33–40 (2013)

    CrossRef  Google Scholar 

  10. Bird, S., Hutchinson, B., Kenthapadi, K., Kıcıman, E., Mitchell, M.: Fairness-aware machine learning: practical challenges and lessons learned. In: Companion Proceedings of the 2019 World Wide Web Conference, pp. 1297–1298 (2019)

    Google Scholar 

  11. Buolamwini, J., Gebru, T.: Gender shades: intersectional accuracy disparities in commercial gender classification. In: Proceedings of the 1st Conference on Fairness, Accountability and Transparency. Proceedings of Machine Learning Research, vol. 81, pp. 77–91. PMLR (2018)

    Google Scholar 

  12. Cao, Q., Shen, L., Xie, W., Parkhi, O.M., Zisserman, A.: VGGFace2: a dataset for recognising faces across pose and age. In: International Conference on Automatic Face Gesture Recognition (FG), pp. 67–74 (2018)

    Google Scholar 

  13. Cavazos, J.G., Phillips, P.J., Castillo, C.D., O’Toole, A.J.: Accuracy comparison across face recognition algorithms: where are we on measuring race bias? CoRR abs/1912.07398 (2019)

    Google Scholar 

  14. Chopra, S., Hadsell, R., LeCun, Y.: Learning a similarity metric discriminatively, with application to face verification. In: Conference on Computer Vision and Pattern Recognition (CVPR), vol. 1, pp. 539–546 (2005)

    Google Scholar 

  15. Chouldechova, A.: Fair prediction with disparate impact: a study of bias in recidivism prediction instruments. Big Data 5(2), 153–163 (2017)

    CrossRef  Google Scholar 

  16. Davies, B., Innes, M., Dawson, A.: An Evaluation of South Wales Police’s Use of Automated Facial Recognition, September 2018. https://static1.squarespace.com/static/51b06364e4b02de2f57fd72e/t/5bfd4fbc21c67c2cdd692fa8/1543327693640/AFR+Report+%5BDigital%5D.pdf. Accessed 5 Aug 2020

  17. Deng, J., Guo, J., Xue, N., Zafeiriou, S.: ArcFace: additive angular margin loss for deep face recognition. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2019)

    Google Scholar 

  18. Dieterich, W., Mendoza, C., Brennan, T.: Compas risk scales: demonstrating accuracy equity and predictive parity performance of the compas risk scales in broward county, July 2016. https://go.volarisgroup.com/rs/430-MBX-989/images/ProPublica_Commentary_Final_070616.pdf. Accessed 5 Aug 2020

  19. Drozdowski, P., Rathgeb, C., Dantcheva, A., Damer, N., Busch, C.: Demographic bias in biometrics: a survey on an emerging challenge. CoRR abs/2003.02488 (2020)

    Google Scholar 

  20. Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.S.: Fairness through awareness. CoRR abs/1104.3913 (2011)

    Google Scholar 

  21. Escalante, H.J., et al.: Modeling, recognizing, and explaining apparent personality from videos. IEEE Trans. Affect. Comput. (2020)

    Google Scholar 

  22. Friedler, S.A., Scheidegger, C., Venkatasubramanian, S., Choudhary, S., Hamilton, E.P., Roth, D.: A comparative study of fairness-enhancing interventions in machine learning. In: Proceedings of the Conference on Fairness, Accountability, and Transparency, pp. 329–338. ACM (2019)

    Google Scholar 

  23. Grother, P., Ngan, M., Hanaoka, K.: Face Recognition Vendor Test (FRVT) Part 3: Demographic Effects. Technical report, National Institute of Standards and Technology (NIST) Interagency/Internal Report (NISTIR) - 8280 (2019)

    Google Scholar 

  24. Guo, G., Zhang, N.: A survey on deep learning based face recognition. Comput. Vis. Image Underst. 189, 102805 (2019)

    CrossRef  Google Scholar 

  25. Guo, Y., Zhang, L., Hu, Y., He, X., Gao, J.: MS-Celeb-1M: a dataset and benchmark for large-scale face recognition. CoRR abs/1607.08221 (2016)

    Google Scholar 

  26. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2016)

    Google Scholar 

  27. Huang, C., Li, Y., Loy, C.C., Tang, X.: Deep imbalanced learning for face recognition and attribute prediction. CoRR abs/1806.00194 (2018)

    Google Scholar 

  28. Huang, Y., et al.: Improving face recognition from hard samples via distribution distillation loss. CoRR abs/2002.03662 (2020)

    Google Scholar 

  29. Jayaraman, U., Gupta, P., Gupta, S., Arora, G., Tiwari, K.: Recent development in face recognition. Neurocomputing 408, 231–245 (2020)

    CrossRef  Google Scholar 

  30. Kemelmacher-Shlizerman, I., Seitz, S.M., Miller, D., Brossard, E.: The megaface benchmark: 1 million faces for recognition at scale. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4873–4882 (2016)

    Google Scholar 

  31. Kleinberg, J., Mullainathan, S., Raghavan, M.: Inherent trade-offs in the fair determination of risk scores. CoRR abs/1609.05807 (2016)

    Google Scholar 

  32. Kortylewski, A., Egger, B., Schneider, A., Gerig, T., Morel-Forster, A., Vetter, T.: Analyzing and reducing the damage of dataset bias to face recognition with synthetic data. In: Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 1–8 (2019)

    Google Scholar 

  33. Kusner, M.J., Loftus, J., Russell, C., Silva, R.: Counterfactual fairness. In: Guyon, I., et al. (eds.) Advances in Neural Information Processing Systems 30, pp. 4066–4076. Curran Associates, Inc. (2017)

    Google Scholar 

  34. Learned-Miller, E., Huang, G.B., RoyChowdhury, A., Li, H., Hua, G.: Labeled faces in the wild: a survey. In: Kawulok, M., Celebi, M.E., Smolka, B. (eds.) Advances in Face Detection and Facial Image Analysis, pp. 189–248. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-25958-1_8

    CrossRef  Google Scholar 

  35. Li, J., et al.: DSFD: dual shot face detector. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2019)

    Google Scholar 

  36. Lo Piano, S.: Ethical principles in machine learning and artificial intelligence: cases from the field and possible ways forward. Humanit. Soc. Sci. Commun. 7(9), 1–7 (2020)

    Google Scholar 

  37. Maze, B., et al.: IARPA Janus benchmark - C: face dataset and protocol. In: International Conference on Biometrics (ICB), pp. 158–165 (2018)

    Google Scholar 

  38. Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., Galstyan, A.: A survey on bias and fairness in machine learning. CoRR abs/1908.09635 (2019)

    Google Scholar 

  39. Merler, M., Ratha, N.K., Feris, R.S., Smith, J.R.: Diversity in faces. CoRR abs/1901.10436 (2019)

    Google Scholar 

  40. Morales, A., Fiérrez, J., Vera-Rodríguez, R.: Sensitivenets: learning agnostic representations with application to face recognition. CoRR abs/1902.00334 (2019)

    Google Scholar 

  41. Nabi, R., Shpitser, I.: Fair inference on outcomes. CoRR abs/1705.10378 (2017)

    Google Scholar 

  42. Pearl, J.: Causal inference in statistics: an overview. Stat. Surv. 3, 96–146 (2009)

    CrossRef  MathSciNet  Google Scholar 

  43. Pessach, D., Shmueli, E.: Algorithmic fairness. CoRR abs/2001.09784 (2020)

    Google Scholar 

  44. Pierce, J., Wong, R.Y., Merrill, N.: Sensor illumination: exploring design qualities and ethical implications of smart cameras and image/video analytics. In: Conference on Human Factors in Computing Systems, pp. 1–19 (2020)

    Google Scholar 

  45. Angwin, J., Larson, J., Mattu, S., Kirchner, L.: Machine bias: there’s software used across the country to predict future criminals and it’s biased against blacks. ProPublica, May 2016. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing. Accessed 5 Aug 2020

  46. Raji, I.D., Gebru, T., Mitchell, M., Buolamwini, J., Lee, J., Denton, E.: Saving face: investigating the ethical concerns of facial recognition auditing. In: Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, pp. 145–151 (2020)

    Google Scholar 

  47. Robinson, J.P., Livitz, G., Henon, Y., Qin, C., Fu, Y., Timoner, S.: Face recognition: too bias, or not too bias? In: Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 1–10 (2020)

    Google Scholar 

  48. Rothe, R., Timofte, R., Gool, L.V.: DEX: deep expectation of apparent age from a single image. In: International Conference on Computer Vision Workshops (ICCVW), pp. 252–257 (2015)

    Google Scholar 

  49. Rothe, R., Timofte, R., Gool, L.V.: Deep expectation of real and apparent age from a single image without facial landmarks. Int. J. Comput. Vision 126(2–4), 144–157 (2018)

    CrossRef  MathSciNet  Google Scholar 

  50. Srinivas, N., Ricanek, K., Michalski, D., Bolme, D.S., King, M.: Face recognition algorithm bias: performance differences on images of children and adults. In: Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 2269–2277 (2019)

    Google Scholar 

  51. Terhörst, P., Kolf, J.N., Damer, N., Kirchbuchner, F., Kuijper, A.: Post-comparison mitigation of demographic bias in face recognition using fair score normalization. CoRR abs/2002.03592 (2020)

    Google Scholar 

  52. Wang, X.: China testing facial-recognition surveillance system in Xinjiang - report. The Guardian, January 2018. https://www.theguardian.com/world/2018/jan/18/china-testing-facial-recognition-surveillance-system-in-xinjiang-report. Accessed 5 Aug 2020

  53. Valentino-DeVries, J.: How the Police Use Facial Recognition, and Where It Falls Short. The New York Times, January 2020. https://www.nytimes.com/2020/01/12/technology/facial-recognition-police.html. Accessed 5 Aug 2020

  54. Mozur, P.: Inside China’s Dystopian Dreams: A.I., Shame and Lots of Cameras. The New York Times, July 2018. https://www.nytimes.com/2018/07/08/business/china-surveillance-technology.html. Accessed 5 Aug 2020

  55. Mozur, P.: One Month, 500,000 Face Scans: How China Is Using A.I. to Profile a Minority. The New York Times, April 2019. https://www.nytimes.com/2019/04/14/technology/china-surveillance-artificial-intelligence-racial-profiling.html. Accessed 5 Aug 2020

  56. Greene, J.: Microsoft won’t sell police its facial-recognition technology, following similar moves by Amazon and IBM. The Washington Post, June 2020. https://www.washingtonpost.com/technology/2020/06/11/microsoft-facial-recognition. Accessed 5 Aug 2020

  57. Krishna, A.: IBM CEO’s Letter to Congress on Racial Justice Reform. THINKPolicy Blog, June 2020. https://www.ibm.com/blogs/policy/facial-recognition-sunset-racial-justice-reforms. Accessed 5 Aug 2020

  58. Torralba, A., Efros, A.A.: Unbiased look at dataset bias. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1521–1528 (2011)

    Google Scholar 

  59. US Day One Blog: We are implementing a one-year moratorium on police use of rekognition, June 2020. https://blog.aboutamazon.com/policy/we-are-implementing-a-one-year-moratorium-on-police-use-of-rekognition. Accessed 5 Aug 2020

  60. Verma, S., Rubin, J.: Fairness definitions explained. In: Proceedings of the International Workshop on Software Fairness, pp. 1–7 (2018)

    Google Scholar 

  61. Vowels, M.J., Camgoz, N.C., Bowden, R.: NestedVAE: isolating common factors via weak supervision. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 9202–9212 (2020)

    Google Scholar 

  62. Wang, H., et al.: CosFace: large margin cosine loss for deep face recognition. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2018)

    Google Scholar 

  63. Wang, M., Deng, W.: Mitigating bias in face recognition using skewness-aware reinforcement learning. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 9322–9331 (2020)

    Google Scholar 

  64. Wang, M., Deng, W., Hu, J., Tao, X., Huang, Y.: Racial faces in-the-wild: reducing racial bias by information maximization adaptation network. CoRR abs/1812.00194 (2018)

    Google Scholar 

  65. Wang, T., Zhao, J., Yatskar, M., Chang, K.W., Ordonez, V.: Balanced datasets are not enough: estimating and mitigating gender bias in deep image representations. In: International Conference on Computer Vision (ICCV), pp. 5310–5319 (2019)

    Google Scholar 

  66. Wang, Z., et al.: Towards fairness in visual recognition: effective strategies for bias mitigation. In: Conference on Computer Vision and Pattern Recognition (CVPR), pp. 8919–8928 (2020)

    Google Scholar 

  67. Yu, J., Hao, X., Xie, H., Yu, Y.: Fair face recognition using data balancing, enhancement and fusion. In: Proceedings of the European Conference on Computer Vision (ECCV) Workshops (ECCVW) (2020, in press)

    Google Scholar 

  68. Yucer, S., Akcay, S., Al-Moubayed, N., Breckon, T.P.: Exploring racial bias within face recognition via per-subject adversarially-enabled data augmentation. In: Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2020)

    Google Scholar 

  69. Zhang, L., Wu, Y., Wu, X.: A causal framework for discovering and removing direct and indirect discrimination. CoRR abs/1611.07509 (2016)

    Google Scholar 

  70. Zhou, S.: AsArcFace: asymmetric additive angular margin loss for fairface recognition. In: Proceedings of the European Conference on Computer Vision Workshops (ECCVW) (2020, in press)

    Google Scholar 

Download references

Acknowledgment

This work has been partially supported by the Spanish projects RTI2018-095232-B-C22 and PID2019-105093GB-I00 (MINECO/FEDER, UE), ICREA under the ICREA Academia programme, and CERCA Programme/Generalitat de Catalunya. We gratefully acknowledge the support of NVIDIA Corporation with the donation of the GPU used for this research.

Author information

Authors and Affiliations

  1. Czech Technical University in Prague, Prague, Czech Republic

    Tomáš Sixta

  2. Universitat Oberta de Catalunya, Barcelona, Spain

    Julio C. S. Jacques Junior

  3. Computer Vision Center, Barcelona, Spain

    Julio C. S. Jacques Junior, Pau Buch-Cardona & Sergio Escalera

  4. Facultat de Matematiques i Informatica, Universitat de Barcelona, Barcelona, Spain

    Pau Buch-Cardona & Sergio Escalera

  5. Anyvision, London, UK

    Eduard Vazquez

Authors
  1. Tomáš Sixta
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Julio C. S. Jacques Junior
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Pau Buch-Cardona
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Eduard Vazquez
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Sergio Escalera
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Tomáš Sixta or Julio C. S. Jacques Junior .

Editor information

Editors and Affiliations

  1. University of Clermont Auvergne, Clermont Ferrand, France

    Prof. Adrien Bartoli

  2. Università degli Studi di Udine, Udine, Italy

    Andrea Fusiello

1 Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 1070 KB)

Rights and permissions

Reprints and Permissions

Copyright information

© 2020 Springer Nature Switzerland AG

About this paper

Verify currency and authenticity via CrossMark

Cite this paper

Sixta, T., Jacques Junior, J.C.S., Buch-Cardona, P., Vazquez, E., Escalera, S. (2020). FairFace Challenge at ECCV 2020: Analyzing Bias in Face Recognition. In: Bartoli, A., Fusiello, A. (eds) Computer Vision – ECCV 2020 Workshops. ECCV 2020. Lecture Notes in Computer Science(), vol 12540. Springer, Cham. https://doi.org/10.1007/978-3-030-65414-6_32

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-65414-6_32

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-65413-9

  • Online ISBN: 978-3-030-65414-6

  • eBook Packages: Computer ScienceComputer Science (R0)