Обзор и тестирование детекторов фронтальных лиц
Калиновский И.А., Спицын В.Г.

Томский политехнический университет(национальный исследовательский университет) (ТПУ), Томск, Россия

Аннотация:
Статья посвящена сравнению разработанного авторами способа обнаружения лиц, основанного на каскаде компактных свёрточных нейронных сетей, с современными детекторами фронтальных лиц. Приведены результаты тестирования 16 алгоритмов на 2 открытых наборах данных, а также замеры скорости их работы. Выводится общая оценка качества алгоритмов.

Ключевые слова :
детектирование лиц, каскадные классификаторы, свёрточные нейронные сети, глубинное обучение.

Цитирование:
Калиновский, И.А. Обзор и тестирование детекторов фронтальных лиц / И.А. Калиновский, В.Г. Спицын // Компьютерная оптика. – 2016. – Т. 40, № 1. – С. 99-111. – DOI: 10.18287/2412-6179-2016-40-1- 99-111.

Литература:

  1. Kalinovskii, I.A. Compact Convolutional Neural Network Cascade for Face Detection / I.A. Kalinovskii, V.G. Spitsyn [Электронный ресурс]. – 2015. – URL: http://arxiv.org/abs/1508.01292.pdf (дата обращения 01.02.2016).
  2. Viola, P. Rapid object detection using a boosted cascade of simple features / P. Viola, M.J. Jones // IEEE Conference on Computer Vision and Pattern Recognition. – 2001. – Vol. 1. – P. 511-518.
  3. Lienhart, R. An extended set of Haar-like features for rapid object detection / R. Lienhart, J. Maydt // IEEE International Conference on Image Processing. – 2002. – Vol. 1. – P. 900-903.
  4. Jain, V. Online domain adaptation of a pre-trained cascade of classifiers / V. Jain, E. Learned-Miller // IEEE Conference on Computer Vision and Pattern Recognition. – 2011. – P. 577-584.
  5. Subburaman, V. Fast bounding box estimation based face detection / V. Subburaman, S. Marcel // European Conference on Computer Vision, Workshop on Face Detection. – 2010. – P. 1-14.
  6. Markuš, N. A method for object detection based on pixel intensity comparisons organized in decision trees / N. Markuš, M. Frljak, I.S. Pandzic, J. Ahlberg, R. Forchheimer // [Электронный ресурс]. – 2013. – URL: http://arxiv.org/abs/1305.4537.pdf (дата обращения 01.02.2016).
  7. Li, J. Learning SURF cascade for fast and accurate object detection / J. Li, Y. Zhang // IEEE Conference on Computer Vision and Pattern Recognition. – 2013. – P. 3468-3475.
  8. Barr, J.R. The effectiveness of face detection algorithms in unconstrained crowd scenes / J.R. Barr, K.W. Bowyer, P.J. Flynn // IEEE Winter Conference on Applications of Computer Vision. – 2014. – P. 1020-1027.
  9. Yang, B. Aggregate channel features for multi-view face detection / B. Yang, J. Yan, Z. Lei, S.Z. Li // IEEE International Joint Conference on Biometrics. – 2014. – P. 1-8.
  10. Mathias, M. Face detection without bells and whistles / M. Mathias, R. Benenson, M. Pedersoli, L. Van Gool // European Conference on Computer Vision. – 2014. – P. 720-735.
  11. Zhang, C. Improving multiview face detection with multi-task deep convolutional neural networks / C. Zhang, Z. Zhang // IEEE Winter Conference on Applications of Computer Vision. – 2014. – P. 1036-1041.
  12. Chen, D. Joint cascade face detection and alignment / D. Chen, S. Ren, Y. Wei, X. Cao, J. Sun // European Conference on Computer Vision. – 2014. – P. 109-122.
  13. Zhu, X. Face detection, pose estimation, and landmark localization in the wild / X. Zhu, D. Ramanan // IEEE Conference on Computer Vision and Pattern Recognition. – 2012. – P. 2879-2886.
  14. Li, H. Efficient boosted exemplar-based face detection / H. Li, Z. Lin, J. Brandt, X. Shen, G. Hua // IEEE Conference on Computer Vision and Pattern Recognition. – 2014. – P. 1843-1850.
  15. Zeiler, M. Visualizing and understanding convolutional networks / M. Zeiler, R. Fergus // European Conference on Computer Vision. – 2014. – P. 818-833.
  16. Szegedy, C. Going deeper with convolutions / C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, A. Rabinovich // [Электронный ресурс]. – 2014. – URL: http://arxiv.org/abs/1409.4842.pdf (дата обращения 01.02.2016).
  17. Garcia, C. Convolutional face finder: A neural architecture for fast and robust face detection / C. Garcia, M. Delakis // IEEE Transactions on Pattern Analysis and Machine Intelligence. – 2004. – P. 1408-1423.
  18. Osadchy, M. Synergistic face detection and pose estimation with energy-based models / M. Osadchy, Y. LeCun, M. Miller // Journal of Machine Learning Research. – 2007. – P. 1197-1215.
  19. Farfade, S.S. Multi-view face detection using deep convolutional neural networks / S.S. Farfade, M. Saberian, L.-J. Li // International Conference on Multimedia Retrieval. – 2015.
  20. Li, H. A Convolutional neural network cascade for face detection / H. Li, Z. Lin, X. Shen, J. Brandt, G. Hua // IEEE Conference on Computer Vision and Pattern Recognition. – 2015. – P. 5325-5334.
  21. Калиновский, И.A. Алгоритм детектирования лиц на видео сверхвысокого разрешения / И.A. Калиновский, В.Г. Спицын // Техническое зрение в системах управления. – 2015. – С. 95-96.
  22. Ko?stinger, M. Annotated Facial Landmarks in the Wild: A Large-scale, real-world database for facial landmark localization / M. Ko?stinger, P. Wohlhart, P.M. Roth, H. Bischof // IEEE International Conference on Computer Vision Workshops. – 2011. – P. 2144-2151.
  23. Vasilache, N. Fast convolutional nets with fbfft: A GPU performance evaluation / N. Vasilache, J. Johnson, M. Mathieu, S. Chintala, S. Piantino, Y. LeCun // [Электронный ресурс]. – 2014. – URL: http://arxiv.org/abs/1412.7580.pdf (дата обращения 01.02.2016).
  24. Ioffe, S. Batch normalization: accelerating deep network training by reducing internal covariate shift / S. Ioffe, C. Szegedy // [Электронный ресурс]. – 2015. – URL: http://arxiv.org/abs/1502.03167.pdf (дата обращения 01.02.2016)
  25. Lee, C.-Y. Deeply-supervised nets / C.-Y. Lee, S. Xie, P. Gallagher, Z. Zhang, Z. Tu // // [Электронный ресурс]. – 2014. – URL: http://arxiv.org/abs/1409.5185.pdf (дата обращения 01.02.2016)
  26. Wolf, L. Face recognition in unconstrained videos with matched background similarity / L. Wolf, T. Hassner, I. Maoz  // IEEE Conference on Computer Vision and Pattern Recognition. – 2014. – P. 529-534.
  27. Калиновский, И.А. Алгоритм обнаружения лиц на основе сверточной нейронной сети / И.А. Калиновский, В.Г. Спицын // Нейрокомпьютеры: разработка и применение. – 2013. – № 10. – С. 48-53.
  28. Le, Q.V. On Optimization Methods for Deep Learning / Q.V. Le, A. Coates, B. Prochnow, A.Y. Ng // International Conference on Machine Learning. – 2011. – P. 265-272.
  29. Ko?stinger, M. Efficient metric learning for real-world face recognition / M. Ko?stinger. – Graz University of Technology: PhD thesis, 2013.
  30. Pham, M.T. Fast training and selection and Haar features using statistics in boosting-based face detection / M.T. Pham, T.J. Cham // IEEE International Conference on Computer Vision. – 2007. – P. 1-7.
  31. Kienzle, W. Face detection: efficient and rank deficient / W. Kienzle, G. Bakir, M. Franz, B. Scholkopf // Advances in Neural Information Processing Systems. – 2005. – P. 673-680.
  32. Jain, V. FDDB: A Benchmark for face detection in unconstrained settings / V. Jain, E. Learned-Miller // Technical Report UM-CS-2010-009. – University of Massachusetts. – 2010.
  33. Yang, B. Fine-grained evaluation on face detection in the wild / B. Yang, J. Yan, Z. Lei, S.Z. Li // IEEE International Conference on Automatic Face and Gesture Recognition. – 2015.
  34. Klare, B.F. Pushing the frontiers of unconstrained face detection and recognition: IARPA Janus Benchmark A. / B.F. Klare, B. Klein, E. Taborsky, A. Blanton, J. Cheney, K. Allen, P. Grother, A. Mah, M. Burge, A.K. Jain // IEEE Conference on Computer Vision and Pattern Recognition. – 2015. – P. 1931-1939.
  35. Davis, J. The relationship between Precision-Recall and ROC curves / J. Davis, M. Goadrich // International Conference on Machine Learning. – 2006. – P. 233-240.
  36. Everingham, M. The PASCAL visual object classes (VOC) challenge / M. Everingham, L.V. Gool, C. Williams, J. Winn, A. Zisserman // International Journal of Computer Vision. – Vol. 88(2). – 2010. – P. 303-338.
  37. Oro, D. Real-time GPU-based face detection in HD video sequences / D. Oro, C. Fernandez, J.R. Saeta, X. Martorell, J. Hernando // IEEE International Conference Computer Vision Workshops. – 2011. – P. 530-537.
  38. Nguyen, T. A software-based dynamic-warp scheduling approach for load-balancing the Viola–Jones face detection algorithm on GPUs / T. Nguyen, D. Hefenbrock, J. Oberg, R. Kastner, S. Baden // Journal of Parallel and Distributed Computing. – Vol. 73(5). – 2013. – P. 677-685.
  39. Sermanet, P. OverFeat: Integrated recognition, localization and detection using convolutional networks / P. Sermanet, D. Eigen, X. Zhang, M. Mathieu, R. Fergus, Y. LeCun // [Электронный ресурс]. – 2013. – URL: http://arxiv.org/abs/1312.6229.pdf (дата обращения 01.02.2016).

© 2009, IPSI RAS
Institution of Russian Academy of Sciences, Image Processing Systems Institute of RAS, Russia, 443001, Samara, Molodogvardeyskaya Street 151; E-mail: ko@smr.ru; Phones: +7 (846) 332-56-22, Fax: +7 (846) 332-56-20