<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">pribor</journal-id><journal-title-group><journal-title xml:lang="ru">Известия высших учебных заведений. Приборостроение</journal-title><trans-title-group xml:lang="en"><trans-title>Journal of Instrument Engineering</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">0021-3454</issn><issn pub-type="epub">2500-0381</issn><publisher><publisher-name>Национальный исследовательский университет ИТМО</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.17586/0021-3454-2022-65-11-842-850</article-id><article-id custom-type="elpub" pub-id-type="custom">pribor-307</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>МАТЕМАТИЧЕСКОЕ И ПРОГРАММНОЕ ОБЕСПЕЧЕНИЕ  ИНФОРМАЦИОННЫХ СИСТЕМ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>MATHEMATICAL AND SOFTWARE SUPPORT  OF INFORMATION SYSTEMS</subject></subj-group></article-categories><title-group><article-title>Применение методов синтеза обучающих данных для распознавания частично скрытых лиц на изображениях</article-title><trans-title-group xml:lang="en"><trans-title>Application of training data  synthesis methods for recognition of partially hidden faces in images</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Летенков</surname><given-names>М. А.</given-names></name><name name-style="western" xml:lang="en"><surname>Letenkov</surname><given-names>M. A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Максим Андреевич Летенков — лаборатория технологий больших данных социокиберфизических систем; мл. научный сотрудник</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Maхim A. Letenkov — St. Petersburg Institute for Informatics and Automation of the RAS, Laboratory of Big Data Technologies in Socio-Cyberphysical Systems; Junior Researcher</p><p>St. Petersburg</p></bio><email xlink:type="simple">letenkovmaksim@yandex.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Яковлев</surname><given-names>Р. Н.</given-names></name><name name-style="western" xml:lang="en"><surname>Iakovlev</surname><given-names>R. N.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Роман Никитич Яковлев — лаборатория технологий больших данных социокиберфизических систем; мл. научный сотрудник</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Roman N. Iakovlev — St. Petersburg Institute for Informatics and Automation of the RAS, Laboratory of Big Data Technologies in Socio-Cyberphysical Systems; Junior Researcher</p><p>St. Petersburg</p></bio><email xlink:type="simple">iakovlev.r@mail.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Маркитантов</surname><given-names>М. В.</given-names></name><name name-style="western" xml:lang="en"><surname>Markitantov</surname><given-names>M. V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Максим Викторович Маркитантов — лаборатория речевых и многомодальных интерфейсов; мл. научный сотрудник</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Maxim V. Markitantov — St. Petersburg Institute for Informatics and Automation of the RAS, Speech and Multimodal Interfaces Laboratory; Junior Researcher</p><p>St. Petersburg</p></bio><email xlink:type="simple">m.markitantov@yandex.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Рюмин</surname><given-names>Д. А.</given-names></name><name name-style="western" xml:lang="en"><surname>Ryumin</surname><given-names>D. A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Дмитрий Александрович Рюмин — канд. техн. наук; лаборатория речевых и многомодальных интерфейсов; ст. научный сотрудник</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Dmitry A. Ryumin — PhD; St. Petersburg Institute for Informatics and Automation of the RAS, Speech and Multimodal Interfaces Laboratory; Senior Researcher</p><p>St. Petersburg</p></bio><email xlink:type="simple">ryumin.d@iias.spb.su</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Карпов</surname><given-names>А. А.</given-names></name><name name-style="western" xml:lang="en"><surname>Karpov</surname><given-names>A. A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Алексей Анатольевич Карпов — д-р техн. наук, доцент; лаборатория речевых и многомодальных интерфейсов; гл. научный сотрудник</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Alexey A. Karpov — Dr. Sci., Associate Professor; St. Petersburg Institute for Informatics and Automation of the RAS, Speech and Multimodal Interfaces Laboratory; Chief Researcher</p><p>St. Petersburg</p></bio><email xlink:type="simple">karpov@iias.spb.su</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Санкт-Петербургский федеральный исследовательский центр Российской академии наук</institution></aff><aff xml:lang="en"><institution>St. Petersburg Federal Research Center of the RAS</institution></aff></aff-alternatives><pub-date pub-type="collection"><year>2022</year></pub-date><pub-date pub-type="epub"><day>03</day><month>12</month><year>2024</year></pub-date><volume>65</volume><issue>11</issue><fpage>842</fpage><lpage>850</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Национальный исследовательский университет ИТМО, 2024</copyright-statement><copyright-year>2024</copyright-year><copyright-holder xml:lang="ru">Национальный исследовательский университет ИТМО</copyright-holder><copyright-holder xml:lang="en">Национальный исследовательский университет ИТМО</copyright-holder><license xlink:href="https://pribor.ifmo.ru/jour/about/submissions#copyrightNotice" xlink:type="simple"><license-p>https://pribor.ifmo.ru/jour/about/submissions#copyrightNotice</license-p></license></permissions><self-uri xlink:href="https://pribor.ifmo.ru/jour/article/view/307">https://pribor.ifmo.ru/jour/article/view/307</self-uri><abstract><p>Для решения проблемы автоматического распознавания лиц людей, использующих такие средства индивидуальной защиты, как медицинская маска, предложен и апробирован новый подход, основанный на применении методов генерации синтетических изображений частично скрытых лиц и модели распознавания лиц ArcFace. Предложена стратегия формирования обучающих наборов данных и получен ряд соответствующих моделей распознавания. Проведена серия экспериментов, направленных на оценку качества предсказаний полученного решения, и установлена зависимость между результирующим качеством предсказаний, реализуемых моделями распознавания, и объемом синтетических изображений в обучающих наборах данных. Согласно результатам экспериментальных исследований, нейросетевые модели, дообученные на наборах данных, в которых объем искусственно синтезированных изображений составляет 40—60 %, демонстрируют более высокие значения показателя точности распознавания, выше 87 % по количественной метрике AAc (Averaged Accuracy). Использование предложенного подхода позволяет значительно улучшить качество распознавания частично скрытых лиц по сравнению с базовым подходом.</p></abstract><trans-abstract xml:lang="en"><p>A new approach to solving the problem of automatic face recognition of people using personal protective equipment such as a medical mask has been proposed and tested. This approach is based on the use of methods of generating synthetic images of partially hidden faces and the face recognition model ArcFace. A strategy for training data sets formation is proposed and a number of corresponding recognition models are derived. A series of experiments aimed at assessing the quality of predictions of the obtained solution are carried out, and a relationship between the resulting quality of predictions implemented by recognition models and the volume of synthetic images in training datasets is established. According to the results of experimental studies, neural network models, further trained on datasets with volume of artificially synthesized images of 40-60%, demonstrate values of recognition accuracy above 87% on the AAc quantitative metric (Average Accuracy). Using the proposed approach makes it possible to significantly improve the quality of recognition of partially hidden faces compared to the basic approach.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>распознавание лиц</kwd><kwd>нейросетевые модели распознавания</kwd><kwd>ArcFace</kwd><kwd>BRAVE-MASKS</kwd><kwd>генерация синтетических изображений</kwd><kwd>средства индивидуальной защиты</kwd><kwd>глубокое обучение</kwd></kwd-group><kwd-group xml:lang="en"><kwd>face recognition</kwd><kwd>neural network recognition models</kwd><kwd>ArcFace</kwd><kwd>BRAVE-MASKS</kwd><kwd>synthetic image generation</kwd><kwd>personal protective equipment</kwd><kwd>deep learning</kwd></kwd-group><funding-group><funding-statement xml:lang="ru">исследование выполнено за счет Российского фонда фундаментальных исследований (проект  № 20-04-60529-вирусы), а также частично в рамках ведущей научной школы (грант № НШ-17.2022.1.6)</funding-statement><funding-statement xml:lang="en">the research was carried out at the expense of the Russian Foundation for Basic Research (project  N 20-04-60529-viruses), as well as partially within the framework of a leading scientific school (grant N NS-17.2022.1.6).</funding-statement></funding-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Zhang K., Zhang Z., Li Z., Qiao Y. Joint face detection and alignment using multitask cascaded convolutional networks // IEEE Signal Processing Letters. 2016. Vol. 23, N 10. P. 1499—1503. DOI: 10.1109/LSP.2016.2603342.</mixed-citation><mixed-citation xml:lang="en">Zhang K., Zhang Z., Li Z., Qiao Y. IEEE Signal Processing Letters, 2016, no. 10(23), pp. 1499–1503, DOI: 10.1109/LSP.2016.2603342.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Zhang F., Fan X., Ai G., Song J., Qin Y., Wu J. Accurate face detection for high performance // arXiv preprint arXiv:1905.01585. 2019. P. 1—9.</mixed-citation><mixed-citation xml:lang="en">Zhang F., Fan X., Ai G., Song J., Qin Y., Wu J. arXiv preprint arXiv:1905.01585, 2019, рр. 1–9.</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Schroff F., Kalenichenko D., Philbin J. Facenet: A unified embedding for face recognition and clustering // Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2015. P. 815—823. DOI: 10.1109/CVPR.2015.7298682.</mixed-citation><mixed-citation xml:lang="en">Schroff F., Kalenichenko D., Philbin J. Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, рр. 815–823, DOI: 10.1109/CVPR.2015.7298682.</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Deng J., Guo J., Xue N., Zafeiriou S. Arcface: Additive angular margin loss for deep face recognition // Proc. of the IEEE/CVF Conf. on Computer Vision and Pattern Recognition. 2019. P. 4690—4699.</mixed-citation><mixed-citation xml:lang="en">Deng J., Guo J., Xue N., Zafeiriou S. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, рр. 4690–4699.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">He Y., Xu D., Wu L., Jian M., Xiang S., Pan C. LFFD: A light and fast face detector for edge devices // arXiv preprint arXiv:1904.10633. 2019. P. 1—10. DOI: 10.48550/arXiv.1904.10633.</mixed-citation><mixed-citation xml:lang="en">He Y., Xu D., Wu L., Jian M., Xiang S., Pan C. arXiv preprint arXiv:1904.10633, 2019, рр. 1–10, DOI: 10.48550/arXiv.1904.10633.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Parkhi O. M., Vedaldi A., Zisserman A. Deep face recognition // British Mashine Vision Conf.: Proc. 2015. P. 1—12. DOI: 10.5244/C.29.41.</mixed-citation><mixed-citation xml:lang="en">Parkhi O. M., Vedaldi A., Zisserman A. Deep face recognition, 2015, рр. 1–12. DOI: 10.5244/C.29.41.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Rab S., Javaid M., Haleem A., Vaishya R. Face masks are new normal after COVID-19 pandemic // Diabetes &amp; Metabolic Syndrome: Clinical Research &amp; Reviews. 2020. Vol. 14, N 6. P. 1617—1619.</mixed-citation><mixed-citation xml:lang="en">Rab S., Javaid M., Haleem A., Vaishya R. Diabetes &amp; Metabolic Syndrome: Clinical Research &amp; Reviews, 2020, no. 6(14), pp. 1617–1619.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Martínez-Díaz Y., Méndez-Vázquez H., Luevano L. S., Nicolás-Díaz M., Chang L., González-Mendoza M. Towards Accurate and Lightweight Masked Face Recognition: an Experimental Evaluation // IEEE Access. 2021. Vol. 10. P. 7341—7353.</mixed-citation><mixed-citation xml:lang="en">Martínez-Díaz Y., Méndez-Vázquez H., Luevano L. S., Nicolás-Díaz M., Chang L., González-Mendoza M. IEEE Access., 2021, vol. 10, рр. 7341–7353.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Anwar A., Raychowdhury A. Masked face recognition for secure authentication // arXiv preprint arXiv:2008.11104. 2020. P. 1—8.</mixed-citation><mixed-citation xml:lang="en">Anwar A., Raychowdhury A. arXiv preprint arXiv:2008.11104, 2020, рр. 1–8.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Cao Q., Shen L., Xie W., Parkhi O. M., Zisserman A. Vggface2: A dataset for recognising faces across pose and age // 13th IEEE Intern. Conf. on Automatic Face &amp; Gesture Recognition (FG 2018). 2018. P. 67—74.</mixed-citation><mixed-citation xml:lang="en">Cao Q., Shen L., Xie W., Parkhi O.M., Zisserman A. 13th IEEE International Conference on Automatic Face &amp; Gesture Recognition (FG 2018), IEEE, 2018, рр. 67–74.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Guo Y., Zhang L., Hu Y., He X., Gao, J. Ms-celeb-1m: A dataset and benchmark for large-scale face recognition // European Conf. on Computer Vision. Cham: Springer, 2016. P. 87—102.</mixed-citation><mixed-citation xml:lang="en">Guo Y., Zhang L., Hu Y., He X., Gao J. European conference on computer vision, Springer, Cham, 2016, рр. 87–102.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Wang Z., Wang G., Huang B., Xiong Z., Hong Q., Wu H., Liang J. Masked face recognition dataset and application // arXiv preprint arXiv:2003.09093. 2020. P. 1—3.</mixed-citation><mixed-citation xml:lang="en">Wang Z., Wang G., Huang B., Xiong Z., Hong Q., Wu H., Liang J. arXiv preprint arXiv:2003.09093, 2020, рр. 1–3.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Liu W., Wen Y., Yu Z., Li M., Raj B., Song L. Sphereface: Deep hypersphere embedding for face recognition // Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2017. P. 212—220.</mixed-citation><mixed-citation xml:lang="en">Liu W., Wen Y., Yu Z., Li M., Raj B., Song L. Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, рр. 212–220.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Wang H., Wang Y., Zhou Z., Ji X., Gong D., Zhou J., Liu W. Cosface: Large margin cosine loss for deep face recognition // Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2018. P. 5265—5274.</mixed-citation><mixed-citation xml:lang="en">Wang H., Wang Y., Zhou Z., Ji X., Gong D., Zhou J., Liu W. Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, рр. 5265–5274</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Rabinovich A. Going deeper with convolutions // Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2015. P. 1—9.</mixed-citation><mixed-citation xml:lang="en">Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Rabinovich A. Going deeper with convolutions // Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2015. P. 1—9.</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Kemelmacher-Shlizerman I. et al. The megaface benchmark: 1 million faces for recognition at scale // Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2016. P. 4873—4882.</mixed-citation><mixed-citation xml:lang="en">Kemelmacher-Shlizerman I. et al. The megaface benchmark: 1 million faces for recognition at scale // Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 2016. P. 4873—4882.</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Letenkov M. A., Iakovlev R. N., Markitantov M. V., Ryumin D. A., Saveliev A. I., Karpov A. A. Method for Generating Synthetic Images of Masked Human Faces // Scientific Visualization. 2022. Vol. 14, N 2. P. 1—17. DOI: 10.26583/sv.14.2.01.</mixed-citation><mixed-citation xml:lang="en">Letenkov M. A., Iakovlev R. N., Markitantov M. V., Ryumin D. A., Saveliev A. I., Karpov A. A. Method for Generating Synthetic Images of Masked Human Faces // Scientific Visualization. 2022. Vol. 14, N 2. P. 1—17. DOI: 10.26583/sv.14.2.01.</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">InsightFace: 2D and 3D Face Analysis Project [Электронный ресурс]: https://github.com/deepinsight/insightface 08.07.2022.</mixed-citation><mixed-citation xml:lang="en">InsightFace: 2D and 3D Face Analysis Project [Электронный ресурс]: https://github.com/deepinsight/insightface 08.07.2022.</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Markitantov M., Ryumina E., Ryumin D., Karpov A. Biometric Russian Audio-Visual Extended MASKS (BRAVEMASKS) Corpus: Multimodal Mask Type Recognition Task // Proc. of ISCA Intern. Conf. INTERSPEECH-2022. Korea. 2022.</mixed-citation><mixed-citation xml:lang="en">Markitantov M., Ryumina E., Ryumin D., Karpov A. Biometric Russian Audio-Visual Extended MASKS (BRAVEMASKS) Corpus: Multimodal Mask Type Recognition Task // Proc. of ISCA Intern. Conf. INTERSPEECH-2022. Korea. 2022.</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
