生物辨識中,原先是由執法單位用於辨認身分,而人臉辨識中的人臉驗證是最不與主體接出且最能秘密進行,一旦取得人臉即能分析並進行認證,此模式已逐漸應用各個場域之中。

以往的人臉驗證研究,訓練模型通常使用大量的訓練資料建立模型,在進行人臉驗證之研究及評估,雖然大量訓練資料有助於穩定及提升人臉驗證辨識率,但較少人著墨於使用少量人臉能判斷人臉驗證之效期,因此本研究探討使用少量人臉資料建立模型並進行人臉驗證研究。

本研究使用少量的人臉,並從人臉照片中取得人臉特徵進行訓練模型,並透過支持向量機(Support Vector Machine,SVM)分別於每個人訓練模型,透過建立人臉模型以進行人臉驗證之方式,提高人臉驗證的表現。

透過本研究所使用支持向量機於不同的訓練資料下建立模型,並以 10 張人臉照片建立支持向量機模型,於人臉驗證中能維持二十一週 80%的辨識率。

In biometric identification, Face Verification were originally used to identify identities by law enforcement agencies. Face Verification can avoid contact with the subjects and complete the tasks privately. Once a face profile is obtained, it can be analyzed and authenticated. This method has been gradually applied in various fields.

In the past, Face Verification often used huge amount of data to verify face profiles. However, although –using lots of data can improve the stability and recognizability of Face Verification, it can ralely use a small amount of face profile data to complete Face Verification tasks. Hence, this research aimed to use –a small amount of face data to build a Face Verification mechanism.

This research proposed a method to extract face features by using feature learning and built the Support Vector Machine (SVM) model to verify a pair of facial images that belong to the same or different subjects. According to research, we can use within 10 face data to train a face model. The result shows that it can maintain an 80% recognition rate in 21 weeks.

[1] Ahonen, T., Hadid, A., & Pietikainen, M. (2006). Face description with local binary patterns: Application to face recognition. IEEE transactions on pattern analysis and machine intelligence, 28(12), 2037-2041.
[2] Chen, J. C., Patel, V. M., & Chellappa, R. (2016, March). Unconstrained face verification using deep cnn features. In 2016 IEEE winter conference on applications of computer vision (WACV) (pp. 1-9). IEEE.
[3] Geitey, A. (2017). Face Recognition. Retrieved from https://face_recognition.readthedocs.io/en/latest/face_recognition.html.
[4] Georghiades, A. S., Belhumeur, P. N., & Kriegman, D. J. (2001). From few to many: Illumination cone models for face recognition under variable lighting and pose. IEEE transactions on pattern analysis and machine intelligence, 23(6), 643-660.
[5] Hu, J., Lu, J., & Tan, Y. P. (2014). Discriminative deep metric learning for face verification in the wild. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1875-1882).
[6] King, D. E. (2009). Dlib-ml: A machine learning toolkit. The Journal of Machine Learning Research, 10, 1755-1758.
[7] Li, H., Hua, G., Lin, Z., Brandt, J., & Yang, J. (2013). Probabilistic elastic matching for pose variant face verification. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 3499-3506).
[8] Lu, J., Hu, J., & Zhou, J. (2017). Deep metric learning for visual understanding: An overview of recent advances. IEEE Signal Processing Magazine, 34(6), 76-84.
[9] Mahalingam, G., & Kambhamettu, C. (2010, December). Face verification with aging using AdaBoost and local binary patterns. In Proceedings of the Seventh Indian Conference on Computer Vision, Graphics and Image Processing (pp. 101-108).
[10] Nguyen, H. V., & Bai, L. (2010, November). Cosine similarity metric learning for face verification. In Asian conference on computer vision (pp. 709-720). Springer, Berlin, Heidelberg.
[11] Olivares-Mercado, J., Aguilar-Torres, G., Toscano-Medina, K., Nakano-Miyatake, M., & Perez-Meana, H. (2011). GMM vs SVM for Face Recognition and Face verification. Reviews, Refinements and New Ideas in Face Recognition, 1-338.
[12] Pereira, J. F., Barreto, R. M., Cavalcanti, G. D., & Tsang, R. (2011, May). A robust feature extraction algorithm based on class-modular image principal component analysis for face verification. In 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 1469-1472). IEEE.
[13] Prabhakar, S., Pankanti, S., & Jain, A. K. (2003). Biometric recognition: Security and privacy concerns. IEEE security & privacy, 1(2), 33-42.
[14] Simonyan, K., Parkhi, O. M., Vedaldi, A., & Zisserman, A. (2013, September). Fisher vector faces in the wild. In BMVC (Vol. 2, No. 3, p. 4).
[15] Sun, Y., Wang, X., & Tang, X. (2013). Hybrid deep learning for face verification. In Proceedings of the IEEE international conference on computer vision (pp. 1489-1496).
[16] Syambas, N. R., & Purwanto, U. H. (2012, October). Image processing and face detection analysis on face verification based on the age stages. In 2012 7th International Conference on Telecommunication Systems, Services, and Applications (TSSA) (pp. 289-293). IEEE.
[17] Taigman, Y., Yang, M., Ranzato, M. A., & Wolf, L. (2014). Deepface: Closing the gap to human-level performance in face verification. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1701-1708).
[18] Thyagharajan, A., & Routray, A. (2017, July). An ensemble metric learning scheme for face recognition. In 2017 IEEE International Conference on Multimedia and Expo (ICME) (pp. 115-120). IEEE.
[19] Tyagi, S., & Khanna, P. (2013). Face verification and identification using DCT-NNDA and SIFT with score-level fusion. International Journal of Biomedical Engineering and Technology, 13(2), 154-176.
[20] Wang, C., Lin, K., & Hung, Y. P. (2014, January). Face verification using LBP feature and clustering. In 2014 International Conference on Computer Vision Theory and Applications (VISAPP) (Vol. 1, pp. 572-578). IEEE.
[21] Xia, Z., Peng, X., Feng, X., & Hadid, A. (2017). Scarce face recognition via two-layer collaborative representation. IET Biometrics, 7(1), 56-62.
[22] Yan, R., Zhong, Z., Zhang, J., & Xu, Y. (2016, December). An improved similarity metric based on joint Bayesian for face verification. In 2016 13th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) (pp. 222-226). IEEE.
[23] You, F., Cao, Y., & Zhang, C. (2017, November). Deep Domain Adaptation with a Few Samples for Face Identification. In 2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) (pp. 178-183). IEEE.
[24] Zhou, L., Wang, H., Lu, Z. M., Nie, T., & Zhao, K. (2016). Face recognition based on LDA and improved pairwise-constrained multiple metric learning method. Journal of Information Hiding and Multimedia Signal Processing, 7(5), 1092.