Learn to Propagate Reliably on Noisy Affinity Graphs

Recent works have shown that exploiting unlabeled data through label propagation can substantially reduce the labeling cost, which has been a critical issue in developing visual recognition models. Yet, how to propagate labels reliably, especially on a dataset with unknown outliers, remains an open question. Conventional methods such as linear diffusion lack the capability of handling complex graph structures and may perform poorly when the seeds are sparse. Latest methods based on graph neural networks would face difficulties on performance drop as they scale out to noisy graphs. To overcome these difficulties, we propose a new framework that allows labels to be propagated reliably on large-scale real-world data. This framework incorporates (1) a local graph neural network to predict accurately on varying local structures while maintaining high scalability, and (2) a confidence-based path scheduler that identifies outliers and moves forward the propagation frontier in a prudent way. Experiments on both ImageNet and Ms-Celeb-1M show that our confidence guided framework can significantly improve the overall accuracies of the propagated labels, especially when the graph is very noisy.

[1]  Dahua Lin,et al.  Person Search in Videos with One Portrait Through Visual and Temporal Links , 2018, ECCV.

[2]  Dong-Hyun Lee,et al.  Pseudo-Label : The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks , 2013 .

[3]  Harri Valpola,et al.  Weight-averaged consistency targets improve semi-supervised deep learning results , 2017, ArXiv.

[4]  Dahua Lin,et al.  Online Multi-modal Person Search in Videos , 2020, ECCV.

[5]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[6]  Bolei Zhou,et al.  A Graph-Based Framework to Bridge Movies and Synopses , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[7]  Max Welling,et al.  Modeling Relational Data with Graph Convolutional Networks , 2017, ESWC.

[8]  Martial Hebert,et al.  Semi-Supervised Self-Training of Object Detection Models , 2005, 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1.

[9]  Zoubin Ghahramani,et al.  Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning , 2015, ICML.

[10]  Lei Yang,et al.  Accelerated Training for Massive Classification via Dynamic Class Selection , 2018, AAAI.

[11]  Dahua Lin,et al.  Caption-Supervised Face Recognition: Training a State-of-the-Art Face Model Without Manual Annotation , 2020, ECCV.

[12]  Honglak Lee,et al.  SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image Editing , 2019, ECCV.

[13]  Pietro Liò,et al.  Graph Attention Networks , 2017, ICLR.

[14]  Yuxiao Hu,et al.  MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition , 2016, ECCV.

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

[16]  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).

[17]  Dahua Lin,et al.  Learning to Cluster Faces via Confidence and Connectivity Estimation , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Prateek Yadav,et al.  Confidence-based Graph Convolutional Networks for Semi-Supervised Learning , 2019, AISTATS.

[19]  Cao Xiao,et al.  FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling , 2018, ICLR.

[20]  Colin Raffel,et al.  Realistic Evaluation of Deep Semi-Supervised Learning Algorithms , 2018, NeurIPS.

[21]  Dahua Lin,et al.  MovieNet: A Holistic Dataset for Movie Understanding , 2020, ECCV.

[22]  Diane J. Cook,et al.  Graph-based anomaly detection , 2003, KDD '03.

[23]  Yu Wang,et al.  Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets , 2020, ECCV.

[24]  VARUN CHANDOLA,et al.  Anomaly detection: A survey , 2009, CSUR.

[25]  Enhua Wu,et al.  Squeeze-and-Excitation Networks , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  Kilian Q. Weinberger,et al.  Simplifying Graph Convolutional Networks , 2019, ICML.

[27]  Dahua Lin,et al.  Placepedia: Comprehensive Place Understanding with Multi-Faceted Annotations , 2020, ECCV.

[28]  Dahua Lin,et al.  Unifying Identification and Context Learning for Person Recognition , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[29]  Zhiyuan Liu,et al.  Graph Neural Networks: A Review of Methods and Applications , 2018, AI Open.

[30]  Stephan Günnemann,et al.  Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking , 2017, ICLR.

[31]  Xiaojin Zhu,et al.  --1 CONTENTS , 2006 .

[32]  Ira Kemelmacher-Shlizerman,et al.  The MegaFace Benchmark: 1 Million Faces for Recognition at Scale , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

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

[34]  Alexander Gammerman,et al.  Transduction with Confidence and Credibility , 1999, IJCAI.

[35]  Felix Naumann,et al.  Data fusion , 2009, CSUR.

[36]  Zoubin Ghahramani,et al.  Learning from labeled and unlabeled data with label propagation , 2002 .

[37]  Bolei Zhou,et al.  A Unified Framework for Shot Type Classification Based on Subject Centric Lens , 2020, ECCV.

[38]  Steven Skiena,et al.  DeepWalk: online learning of social representations , 2014, KDD.

[39]  Jure Leskovec,et al.  Inductive Representation Learning on Large Graphs , 2017, NIPS.

[40]  Junjie Yan,et al.  Consensus-Driven Propagation in Massive Unlabeled Data for Face Recognition , 2018, ECCV.

[41]  Yannis Avrithis,et al.  Label Propagation for Deep Semi-Supervised Learning , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[42]  Bolei Zhou,et al.  A Local-to-Global Approach to Multi-Modal Movie Scene Segmentation , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[43]  Bernhard Schölkopf,et al.  Learning with Local and Global Consistency , 2003, NIPS.

[44]  Charles Blundell,et al.  Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles , 2016, NIPS.

[45]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[46]  Heng-Tze Cheng,et al.  Wide & Deep Learning for Recommender Systems , 2016, DLRS@RecSys.

[47]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[48]  Mikhail Belkin,et al.  Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples , 2006, J. Mach. Learn. Res..

[49]  Dahua Lin,et al.  From Trailers to Storylines: An Efficient Way to Learn from Movies , 2018, ArXiv.

[50]  Lei Yang,et al.  Learning to Cluster Faces on an Affinity Graph , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).