Face Feature Extraction and Recognition Based on Lighted Deep Convolutional Neural Network

Face recognition is one of the most challenging research fields in image processing, pattern recognition and artificial intelligence, and its practical application has a bright future. Face is vulnerable to illumination and expression factors in the process of image acquisition. The traditional machine learning method for extracting face features usually requires high clarity of face images to achieve ideal results. Face feature extraction and recognition based on deep learning method can effectively solve this kind of problem, but the recognition rate is greatly improved, at the same time, the size of the network and the difficulty of training are also sharply increased. It does not meet the increasingly lightweight requirements of the network. This paper studies the current mainstream DCNN architecture, especially MobileNet and ShuffleNet at light-weight mobile level. Some high-efficiency and practical structural features of convolution are summarized; A Nonlinear separable deep convolutional neural network X-Net is designed; Focusing on the loss of the sample in the real face dataset, focus loss is introduced to reduce the effect of sample deviation on network fitting. The experimental results show that the feature extraction based on deep convolutional neural network in this paper can use dynamic library DLL to play an effective role in practical commercial projects.

[1]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Graham W. Taylor,et al.  Adaptive deconvolutional networks for mid and high level feature learning , 2011, 2011 International Conference on Computer Vision.

[3]  Bo Chen,et al.  MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications , 2017, ArXiv.

[4]  Xiaogang Wang,et al.  Deep Learning Face Representation by Joint Identification-Verification , 2014, NIPS.

[5]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[6]  Luo Siwei,et al.  A Novel Regularization Method Based on Convolution Neural Network , 2014 .

[7]  Wen Gao,et al.  Learned local Gabor patterns for face representation and recognition , 2009, Signal Process..

[8]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[9]  Kate Smith-Miles,et al.  Context-aware fusion: A case study on fusion of gait and face for human identification in video , 2010, Pattern Recognit..

[10]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[11]  Xiaogang Wang,et al.  Deep Learning Face Representation from Predicting 10,000 Classes , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Xiangyu Zhang,et al.  ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[13]  Ming Yang,et al.  DeepFace: Closing the Gap to Human-Level Performance in Face Verification , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[14]  James Philbin,et al.  FaceNet: A unified embedding for face recognition and clustering , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  Kaiming He,et al.  Focal Loss for Dense Object Detection , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).