(46-5) 14 * << * >> * Russian * English * Content * All Issues
Open-set face identification with automatic detection of out-of-distribution images
A.D. Sokolova 1, A.V. Savchenko 1, S.I. Nikolenko 2
1 National Research University Higher School of Economics,
603093, Nizhny Novgorod, Russia, Rodionova 136;
2 Saint Petersburg University, 199034, Saint-Petersburg, Russia, Universitetskaya nab. 7-9
PDF, 829 kB
Full text of article: Russian language.
One of main issues in face identification is the lack of training data of specific type (bad quality image, varying scale or illumination, children/old people faces, etc.). As a result, the recogni-tion accuracy may be low for input images which are not similar to the majority of images in the dataset used to train the feature extractor. In this paper, we propose that this issue is dealt with by the automatic detection of such out-of-distribution data based on the addition of a preliminary stage of their automatic rejection using a special convolutional network trained using a set of rare data collected using various transformations. To increase the computational efficiency, the decision about the presence of a rare image is made on the basis of the same face descriptor that is used in the classifier. Experimental research confirmed the accuracy improvement of the proposed approach for several datasets of faces and modern neural network descriptors.
face recognition, anomaly detection, image processing, detection of out-of-distribution images.
Sokolova AD, Savchenko AV, Nikolenko SI. Open-set face identification with automatic detection of out-of-distribution images. Computer Optics 2022; 46(5): 801-807. DOI: 10.18287/2412-6179-CO-1061.
The research was supported by the Russian Science Foundation under RSF grant 20-71-10010. The research work by S. Nikolenko was supported by St. Petersburg State University under project # 73555239 "Artificial Intelligence and Data Science: Theory, Technology, Industrial and Interdisciplinary Research and Applications''.
- Deng J, Guo J, Xue N, Zafeiriou S. ArcFace: Additive angular margin loss for deep face. 2019 IEEE/CVF Conf on Computer Vision and Pattern Recognition (CVPR) 2019: 4690-4699.
- Cao Q, Shen L, Xie W, Parkhi OM, Zisserman A. Vggface2: A dataset for recognizing faces across pose and age. 13th IEEE Int Conf on Automatic Face & Gesture Recognition (FG 2018) 2018: 67-74.
- Guo Y, Zhang L, Hu Y, He X, Gao J. MS-Celeb-1M: A dataset and benchmark for large-scale face recognition. In Book: Leibe B, Matas J, Sebe N, Welling M, eds. Computer Vision – ECCV 2016. Cham: Springer; 2016: 87-102.
- Lahasan B, Lutfi SL, San-Segundo R. A survey on techniques to handle face recognition challenges: occlusion, single sample per subject and expression. Artif Intell Rev 2019; 52(2): 949-979.
- Nikitin MYu, Konushin VS, Konushin AS. Neural network model for video-based face recognition with frames quality assessment. Computer Optics 2017, 41(5): 732-742.
- Gunther M, Cruz S, Rudd EM, Boult TE. Toward open-set face recognition. 2017 IEEE Conf on Computer Vision and Pattern Recognition Workshops (CVPRW) 2017: 573-582.
- Xie H, Du Y, Yu H, Chang Y, Xu Z, Tang Y. Open set face recognition with deep transfer learning and extreme value statistics. Int J Wavelets Multiresolut Inf Process 2018; 16(4): 1850034.
- Savchenko AV. Efficient facial representations for age, gender and identity recognition in organizing photo albums using multi-output ConvNet. PeerJ Comput Sci 2019; 5: e197.
- Sokolova AD, Savchenko AV. Computation-efficient face recognition algorithm using a sequential analysis of high dimensional neural-net features. Optical Memory and Neural Networks 2020; 29(1): 19-29.
- Faizov BV, Shakhuro VI, Sanzharov VV, Konushin AS. Classification of rare road signs. Computer Optics 2020, 44(2): 236-243.
- Savchenko AV, Belova NS. Unconstrained face identification using maximum likelihood of distances between deep off-the-shelf features. Expert Syst Appl 2018; 108: 170-182.
- Yu C, Zhu X, Lei Z, Li SZ. Out-of-distribution detection for reliable face recognition. IEEE Signal Process Lett 2020; 27: 710-714.
- Zhou E, Cao Z, Yin Q. Naive-deep face recognition: Touching the limit of LFW benchmark or not? arXiv Preprint. Source: <https:/arxiv.org/abs/1501.04690>.
- Gupta D, Ahmad M. An efficient method to get improved peak signal to noise ratio (PSNR), using support vector machine. Int J Emerg Technol Adv Eng 2017; 7(9): 49-53.
- Hore A, Ziou D. Image quality metrics: PSNR vs. SSIM. Proc 20th Int Conf on Pattern Recognition (ICPR) 2010: 2366-2369.
- Rezende DJ, Mohamed S, Wierstra D. Stochastic backpropagation and approximate inference in deep generative models. Proc 31st Int Conf on Machine Learning (PMLR) 2014: 1278-1286.
- Li KL, Huang HK, Tian SF, Xu W. Improving one-class SVM for anomaly detection. Proc Int Conf on Machine Learning and Cybernetics 2003; 5: 3077-3081.
- Chalapathy R, Menon AK, Chawla S. Anomaly detection using one-class neural networks. arXiv Preprint. Source: <https:/arxiv.org/abs/1802.06360>.
- Hara S, Nitanda A, Maehara T. Data cleansing for models trained with sgd. arXiv Preprint. Source: <https:/arxiv.org/abs/1906.08473>.
- Sokolova AD, Savchenko AV. Organizing data in video surveillance systems based on deep learning technologies. In Book: Proceedings of the IV International Conference "Information Technologies and Nanotechnologies" (ITNT 2018). Samara: "Novaja Tehnika" Publisher; 2018: 946-952.
- Hendrycks D, Gimpel K. A baseline for detecting misclassified and out-of-distribution examples in neural networks. arXiv Preprint. Source: <https:/arxiv.org/abs/1610.02136>.
- Lee K, Lee K, Lee H, Shin J. A simple unified framework for detecting out-of-distribution samples and adversarial attacks. arXiv Preprint. Source: <https:/arxiv.org/abs/1807.03888>.
- Lee K, Lee K, Lee H, Shin J. Training confidence-calibrated classifiers for detecting out-of-distribution samples. arXiv Preprint. Source: <https:/arxiv.org/abs/1711.09325>.
- Huang GB, Mattar M, Berg T, Learned-Miller E. Labeled faces in the wild: A database for studying face recognition in unconstrained environments. Workshop on faces in 'Real-Life' Images: detection, alignment, and recognition, 2008: 1-14. Source: <https://hal.inria.fr/inria-00321923/document>.
- Klare BF, Klein B, Taborsky E, Blanton A, Cheney J, Allen K, Jain AK. Pushing the frontiers of unconstrained face detection and recognition: Iarpa janus benchmark a. 2015 IEEE Conf on Computer Vision and Pattern Recognition (CVPR) 2015: 1931-1939.
- Wang M, Deng W, Hu J, Tao X, Huang Y. Racial faces in the wild: Reducing racial bias by information maximization adaptation network. 2019 IEEE/CVF Int Conf on Computer Vision (ICCV) 2019: 692-702.
- Cheng J, Li Y, Wang J, Yu L, Wang S. Exploiting effective facial patches for robust gender recognition. Tsinghua Sci Technol 2019; 24(3): 333-345.
- Bianco S. Large age-gap face verification by feature injection in deep networks. Pattern Recognit Lett 2017; 90: 36-42.
- Nikolenko SI. Synthetic data for deep learning. Cham: Springer Nature Switzerland AG; 2021.
- Chen YC, Lin H, Shu M, Li R, Tao X, Shen X, Jia J. Facelet-bank for fast portrait manipulation. 2018 IEEE/CVF Conf on Computer Vision and Pattern Recognition 2018: 3541-3549.
- Zhang H, Chen W, He H, Jin Y. Disentangled makeup transfer with generative adversarial network. arXiv Preprint. Source: <https:/arxiv.org/abs/1907.01144>.
- Kharchevnikova AS, Savchenko AV. Neural networks in video-based age and gender recognition on mobile platforms. Opt Mem Neural Networks 2018; 27(4): 246-259.
© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: firstname.lastname@example.org ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20