Improving Face Recognition from Hard Samples via Distribution Distillation Loss

Large facial variations are the main challenge in face recognition. To this end, previous variation-specific methods make full use of task-related prior to design special network losses, which are typically not general among different tasks and scenarios. In contrast, the existing generic methods focus on improving the feature discriminability to minimize the intra-class distance while maximizing the interclass distance, which perform well on easy samples but fail on hard samples. To improve the performance on those hard samples for general tasks, we propose a novel Distribution Distillation Loss to narrow the performance gap between easy and hard samples, which is a simple, effective and generic for various types of facial variations. Specifically, we first adopt state-of-the-art classifiers such as ArcFace to construct two similarity distributions: teacher distribution from easy samples and student distribution from hard samples. Then, we propose a novel distribution-driven loss to constrain the student distribution to approximate the teacher distribution, which thus leads to smaller overlap between the positive and negative pairs in the student distribution. We have conducted extensive experiments on both generic large-scale face benchmarks and benchmarks with diverse variations on race, resolution and pose. The quantitative results demonstrate the superiority of our method over strong baselines, e.g., Arcface and Cosface.

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

[2]  Stefanos Zafeiriou,et al.  UV-GAN: Adversarial Facial UV Map Completion for Pose-Invariant Face Recognition , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

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

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

[5]  Li Shen,et al.  Comparator Networks , 2018, ECCV.

[6]  Zheng Xu,et al.  Training Shallow and Thin Networks for Acceleration via Knowledge Distillation with Conditional Adversarial Networks , 2017, ICLR.

[7]  Anil K. Jain,et al.  IARPA Janus Benchmark-B Face Dataset , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[8]  Rui Zhang,et al.  KDGAN: Knowledge Distillation with Generative Adversarial Networks , 2018, NeurIPS.

[9]  Matti Pietikäinen,et al.  Local frequency descriptor for low-resolution face recognition , 2011, Face and Gesture 2011.

[10]  Xiaoming Liu,et al.  Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition , 2017, IEEE Transactions on Image Processing.

[11]  Sixue Gong,et al.  Jointly De-Biasing Face Recognition and Demographic Attribute Estimation , 2019, ECCV.

[12]  Meng Yang,et al.  Large-Margin Softmax Loss for Convolutional Neural Networks , 2016, ICML.

[13]  Yu Qiao,et al.  A Discriminative Feature Learning Approach for Deep Face Recognition , 2016, ECCV.

[14]  Tieniu Tan,et al.  A Light CNN for Deep Face Representation With Noisy Labels , 2015, IEEE Transactions on Information Forensics and Security.

[15]  Wei Liu,et al.  Super-Identity Convolutional Neural Network for Face Hallucination , 2018, ECCV.

[16]  Jian Yang,et al.  MemNet: A Persistent Memory Network for Image Restoration , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[17]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[18]  Yu Qiao,et al.  Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks , 2016, IEEE Signal Processing Letters.

[19]  Yan Lu,et al.  Relational Knowledge Distillation , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  Xudong Jiang,et al.  Deep Coupled ResNet for Low-Resolution Face Recognition , 2018, IEEE Signal Processing Letters.

[21]  Stefanos Zafeiriou,et al.  ArcFace: Additive Angular Margin Loss for Deep Face Recognition , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Pong C. Yuen,et al.  Very low resolution face recognition problem , 2010, 2010 Fourth IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS).

[23]  Xing Ji,et al.  CosFace: Large Margin Cosine Loss for Deep Face Recognition , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

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

[25]  Pablo H. Hennings-Yeomans,et al.  Simultaneous super-resolution and feature extraction for recognition of low-resolution faces , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[26]  Jian Yang,et al.  Face Recognition With Pose Variations and Misalignment via Orthogonal Procrustes Regression , 2016, IEEE Transactions on Image Processing.

[27]  Xiaoming Liu,et al.  Coefficients Pose-Variant Input Recogni 8 on Engine Frontalized Output Generator FF-GAN D Discriminator Extreme Pose Input Frontalized Output , 2017 .

[28]  Shengcai Liao,et al.  Learning Face Representation from Scratch , 2014, ArXiv.

[29]  Jian Yang,et al.  FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[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]  Rama Chellappa,et al.  Synthesis-based recognition of low resolution faces , 2011, 2011 International Joint Conference on Biometrics (IJCB).

[32]  Greg Mori,et al.  Similarity-Preserving Knowledge Distillation , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[33]  Stefanos Zafeiriou,et al.  Marginal Loss for Deep Face Recognition , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

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

[35]  Hailin Shi,et al.  Co-Mining: Deep Face Recognition With Noisy Labels , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[36]  Qingmin Liao,et al.  Discriminative Multidimensional Scaling for Low-Resolution Face Recognition , 2018, IEEE Signal Processing Letters.

[37]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[38]  Mislav Grgic,et al.  SCface – surveillance cameras face database , 2011, Multimedia Tools and Applications.

[39]  Cheng Li,et al.  Pose-Robust Face Recognition via Deep Residual Equivariant Mapping , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[40]  Shifeng Zhang,et al.  Support Vector Guided Softmax Loss for Face Recognition , 2018, ArXiv.

[41]  Anil K. Jain,et al.  IARPA Janus Benchmark - C: Face Dataset and Protocol , 2018, 2018 International Conference on Biometrics (ICB).

[42]  James M. Rehg,et al.  Fine-Grained Head Pose Estimation Without Keypoints , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[43]  Andrew Zisserman,et al.  Multicolumn Networks for Face Recognition , 2018, BMVC.

[44]  Abhinav Gupta,et al.  Training Region-Based Object Detectors with Online Hard Example Mining , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[45]  Jian Yang,et al.  Image Super-Resolution via Deep Recursive Residual Network , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[46]  Huchuan Lu,et al.  Deep Mutual Learning , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[47]  Ross B. Girshick,et al.  Focal Loss for Dense Object Detection , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[48]  Feiyue Huang,et al.  CurricularFace: Adaptive Curriculum Learning Loss for Deep Face Recognition , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[49]  Omkar M. Parkhi,et al.  VGGFace2: A Dataset for Recognising Faces across Pose and Age , 2017, 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018).

[50]  Jian Cheng,et al.  NormFace: L2 Hypersphere Embedding for Face Verification , 2017, ACM Multimedia.

[51]  Victor S. Lempitsky,et al.  Learning Deep Embeddings with Histogram Loss , 2016, NIPS.

[52]  Dimitris N. Metaxas,et al.  Reconstruction-Based Disentanglement for Pose-Invariant Face Recognition , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[53]  Xiao Zhang,et al.  Range Loss for Deep Face Recognition with Long-Tailed Training Data , 2016, 2017 IEEE International Conference on Computer Vision (ICCV).

[54]  Qijun Zhao,et al.  Towards resolution invariant face recognition in uncontrolled scenarios , 2016, 2016 International Conference on Biometrics (ICB).

[55]  Yu Cheng,et al.  3D-Aided Deep Pose-Invariant Face Recognition , 2018, IJCAI.

[56]  Naiyan Wang,et al.  Like What You Like: Knowledge Distill via Neuron Selectivity Transfer , 2017, ArXiv.

[57]  Shiguang Shan,et al.  A Benchmark and Comparative Study of Video-Based Face Recognition on COX Face Database , 2015, IEEE Transactions on Image Processing.

[58]  Bhiksha Raj,et al.  SphereFace: Deep Hypersphere Embedding for Face Recognition , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[59]  Jian Cheng,et al.  Additive Margin Softmax for Face Verification , 2018, IEEE Signal Processing Letters.

[60]  Xiaoming Liu,et al.  Disentangled Representation Learning GAN for Pose-Invariant Face Recognition , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).