Small Sample Face Recognition Algorithm based on Novel Siamese Network = Small Sample Face Recognition Algorithm based on Novel Siamese Network

다국어 초록 (Multilingual Abstract)

In face recognition, sometimes the number of available training samples for single category is insufficient. Therefore, the performances of models trained by convolutional neural network are not ideal. The small sample face recognition algorithm based on novel Siamese network is proposed in this paper, which doesn’t need rich samples for training. The algorithm designs and realizes a new Siamese network model, SiameseFace1, which uses pairs of face images as inputs and maps them to target space so that the L2 norm distance in target space can represent the semantic distance in input space. The mapping is represented by the neural network in supervised learning. Moreover, a more lightweight Siamese network model, SiameseFace2, is designed to reduce the network parameters without losing accuracy. We also present a new method to generate training data and expand the number of training samples for single category in AR and labeled faces in the wild (LFW) datasets, which improves the recognition accuracy of the models. Four loss functions are adopted to carry out experiments on AR and LFW datasets. The results show that the contrastive loss function combined with new Siamese network model in this paper can effectively improve the accuracy of face recognition.

참고문헌 (Reference)

1 Parchami. M, "Video-based face recognition using ensemble of haar-like deep convolutional neural networks" 2017

2 이상걸, "Variations of AlexNet and GoogLeNet to Improve Korean Character Recognition Performance" 한국정보처리학회 14 (14): 205-217, 2018

3 Tsalakanidou. F, "Use of depth and colour eigenfaces for face recognition" 24 (24): 1427-1435, 2003

4 Blanco-Gonzalo R, "Time evolution of face recognition in accessible scenarios" 5 (5): 24-24, 2015

5 Zhang. L, "Sparse representation or collaborative representation: Which helps face recognition?" 2011

6 Ya. Tu, "Semi-supervised Learning with Generative Adversarial Networks on Digital Signal Modulation Classification" 55 (55): 243-254, 2018

7 Wright. J, "Robust face recognition via sparse representation" 31 (31): 210-227, 2009

8 Vo. D. M, "Robust face recognition via hierarchical collaborative representation" 432 : 332–346-332–346, 2018

9 Zhu. Z, "Recover canonical-view faces in the wild with deep neural networks"

10 Sainath. T. N, "Multichannel signal processing with deep neural networks for automatic speech recognition" 25 (25): 965-979, 2017

11 Liu. F, "Local similarity based linear discriminant analysis for face recognition with single sample per person" 2014

12 Shaham. U, "Learning by coincidence: siamese networks and common variable learning" 74 : 52-63, 2018

13 Chopra. S, "Learning a similarity metric discriminatively, with application to face verification" 2005

14 Koo K M, "Image recognition performance enhancements using image normalization" 7 (7): 33-33, 2017

15 Maze. B, "IARPA Janus Benchmark–C: Face Dataset and Protocol" 2018

16 Whitelam. C, "IARPA Janus Benchmark-B Face Dataset" 2017

17 Yang. M, "Gabor feature based sparse representation for face recognition with gabor occlusion dictionary" 2010

18 Simonyan. K, "Fisher Vector Faces in the Wild" 2013

19 Barkan. O, "Fast high dimensional vector multiplication face recognition" 2013

20 Stephen. ID, "Facial Shape Analysis Identifies Valid Cues to Aspects of Physiological Health in Caucasian, Asian, and African Populations" 8 : 2017

21 Schroff. F, "FaceNet: A unified embedding for face recognition and clustering" 2015

22 He. X, "Face recognition using laplacianfaces" 27 (27): 328-340, 2005

23 Chen Li, "Face Recognition Based on the Combination of Enhanced Local Texture Feature and DBN under Complex Illumination Conditions" 한국정보처리학회 14 (14): 191-204, 2018

24 Tran. L, "Disentangled representation learning gan for pose-invariant face recognition" 2017

25 Taigman. Y, "DeepFace: closing the gap to human-level performance in face verification" 2014

26 Yi. Sun, "Deep learning face representation from predicting 10,000 classes" 2014

27 Yi. Sun, "Deep learning face representation by joint identification-verification" 2014

28 Ning Yu, "Deep Learning in Genomic and Medical Image Data Analysis: Challenges and Approaches" 한국정보처리학회 13 (13): 204-214, 2017

29 Berlemont. S, "Class-balanced siamese neural networks" 273 : 47-56, 2018

30 CHEN. Dong, "Blessing of dimensionality: high dimensional feature and its efficient compression for face verification" 2013

31 Chen. D, "Bayesian face revisited: A joint formulation" 2012