Sparsifying Neural Network Connections for Face Recognition

This paper proposes to learn high-performance deep ConvNets with sparse neural connections, referred to as sparse ConvNets, for face recognition. The sparse ConvNets are learned in an iterative way, each time one additional layer is sparsified and the entire model is re-trained given the initial weights learned in previous iterations. One important finding is that directly training the sparse ConvNet from scratch failed to find good solutions for face recognition, while using a previously learned denser model to properly initialize a sparser model is critical to continue learning effective features for face recognition. This paper also proposes a new neural correlation-based weight selection criterion and empirically verifies its effectiveness in selecting informative connections from previously learned models in each iteration. When taking a moderately sparse structure (26%-76% of weights in the dense model), the proposed sparse ConvNet model significantly improves the face recognition performance of the previous state-of-the-art DeepID2+ models given the same training data, while it keeps the performance of the baseline model with only 12% of the original parameters.

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  Geoffrey E. Hinton,et al.  Distilling the Knowledge in a Neural Network , 2015, ArXiv.

[3]  Joan Bruna,et al.  Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation , 2014, NIPS.

[4]  Rich Caruana,et al.  Do Deep Nets Really Need to be Deep? , 2013, NIPS.

[5]  Babak Hassibi,et al.  Second Order Derivatives for Network Pruning: Optimal Brain Surgeon , 1992, NIPS.

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

[7]  Peter M. Williams,et al.  Bayesian Regularization and Pruning Using a Laplace Prior , 1995, Neural Computation.

[8]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

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

[10]  Jian Sun,et al.  A Practical Transfer Learning Algorithm for Face Verification , 2013, 2013 IEEE International Conference on Computer Vision.

[11]  Andrew Zisserman,et al.  Deep Face Recognition , 2015, BMVC.

[12]  Nitish Srivastava,et al.  Improving neural networks by preventing co-adaptation of feature detectors , 2012, ArXiv.

[13]  Julien Cornebise,et al.  Weight Uncertainty in Neural Networks , 2015, ArXiv.

[14]  Jian Sun,et al.  Bayesian Face Revisited: A Joint Formulation , 2012, ECCV.

[15]  Julien Cornebise,et al.  Weight Uncertainty in Neural Network , 2015, ICML.

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

[17]  Xiaogang Wang,et al.  Deeply learned face representations are sparse, selective, and robust , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Pushmeet Kohli,et al.  Memory Bounded Deep Convolutional Networks , 2014, ArXiv.

[19]  Yann LeCun,et al.  Optimal Brain Damage , 1989, NIPS.

[20]  Anil K. Jain,et al.  Unconstrained Face Recognition: Identifying a Person of Interest From a Media Collection , 2014, IEEE Transactions on Information Forensics and Security.

[21]  Anil K. Jain,et al.  Open source biometric recognition , 2013, 2013 IEEE Sixth International Conference on Biometrics: Theory, Applications and Systems (BTAS).

[22]  Jian Sun,et al.  Accelerating Very Deep Convolutional Networks for Classification and Detection , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[23]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[24]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

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

[26]  Peter N. Belhumeur,et al.  Tom-vs-Pete Classifiers and Identity-Preserving Alignment for Face Verification , 2012, BMVC.

[27]  Ming Yang,et al.  Web-scale training for face identification , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[28]  Andrew Zisserman,et al.  Speeding up Convolutional Neural Networks with Low Rank Expansions , 2014, BMVC.

[29]  R. Venkatesh Babu,et al.  Data-free Parameter Pruning for Deep Neural Networks , 2015, BMVC.

[30]  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.

[31]  Jian Sun,et al.  Blessing of Dimensionality: High-Dimensional Feature and Its Efficient Compression for Face Verification , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[32]  Marwan Mattar,et al.  Labeled Faces in the Wild: A Database forStudying Face Recognition in Unconstrained Environments , 2008 .

[33]  Andrew Zisserman,et al.  Fisher Vector Faces in the Wild , 2013, BMVC.

[34]  Anil K. Jain,et al.  Pushing the frontiers of unconstrained face detection and recognition: IARPA Janus Benchmark A , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[35]  Aditya Bhaskara,et al.  Provable Bounds for Learning Some Deep Representations , 2013, ICML.

[36]  Tal Hassner,et al.  Face recognition in unconstrained videos with matched background similarity , 2011, CVPR 2011.