Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification

Abstract

Person re-identification (re-ID) aims at identifying the same persons' images across different cameras. However, domain diversities between different datasets pose an evident challenge for adapting the re-ID model trained on one dataset to another one. State-of-the-art unsupervised domain adaptation methods for person re-ID transferred the learned knowledge from the source domain by optimizing with pseudo labels created by clustering algorithms on the target domain. Although they achieved state-of-the-art performances, the inevitable label noise caused by the clustering procedure was ignored. Such noisy pseudo labels substantially hinders the model's capability on further improving feature representations on the target domain. In order to mitigate the effects of noisy pseudo labels, we propose to softly refine the pseudo labels in the target domain by proposing an unsupervised framework, Mutual Mean-Teaching (MMT), to learn better features from the target domain via off-line refined hard pseudo labels and on-line refined soft pseudo labels in an alternative training manner. In addition, the common practice is to adopt both the classification loss and the triplet loss jointly for achieving optimal performances in person re-ID models. However, conventional triplet loss cannot work with softly refined labels. To solve this problem, a novel soft softmax-triplet loss is proposed to support learning with soft pseudo triplet labels for achieving the optimal domain adaptation performance. The proposed MMT framework achieves considerable improvements of 14.4%, 18.2%, 13.1% and 16.4% mAP on Market-to-Duke, Duke-to-Market, Market-to-MSMT and Duke-to-MSMT unsupervised domain adaptation tasks. Code is available at https://github.com/yxgeee/MMT.

Accepted in International Conference on Learning Representations (ICLR 2020)

References in corpus (10)

• Distilling the Knowledge in a Neural Network
• Deep Domain Confusion: Maximizing for Domain Invariance
• When Does Label Smoothing Help?
• Domain Separation Networks
• Adversarial Discriminative Domain Adaptation
• Mind the Class Weight Bias: Weighted Maximum Mean Discrepancy for Unsupervised Domain Adaptation
• Unsupervised Person Re-identification by Soft Multilabel Learning
• Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation
• Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-identification
• A Novel Unsupervised Camera-aware Domain Adaptation Framework for Person Re-identification

Cited by in corpus (78)

• Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID
• FastReID: A Pytorch Toolbox for General Instance Re-identification
• Cluster-guided Asymmetric Contrastive Learning for Unsupervised Person Re-Identification
• Hybrid Dynamic Contrast and Probability Distillation for Unsupervised Person Re-Id
• Complementary Pseudo Labels For Unsupervised Domain Adaptation On Person Re-identification
• Learning to Disentangle Scenes for Person Re-identification
• Cluster Contrast for Unsupervised Person Re-Identification
• Unsupervised Person Re-identification with Stochastic Training Strategy
• Self-Supervised Pre-Training for Transformer-Based Person Re-Identification
• Learning Invariance from Generated Variance for Unsupervised Person Re-identification
• Aligning Correlation Information for Domain Adaptation in Action Recognition
• Enhancing Diversity in Teacher-Student Networks via Asymmetric branches for Unsupervised Person Re-identification
• Towards Better Accuracy-efficiency Trade-offs: Divide and Co-training
• CA-Jaccard: Camera-aware Jaccard Distance for Person Re-identification
• Robust Local Preserving and Global Aligning Network for Adversarial Domain Adaptation
• Test-time Batch Statistics Calibration for Covariate Shift
• DomainMix: Learning Generalizable Person Re-Identification Without Human Annotations
• Exploiting Sample Uncertainty for Domain Adaptive Person Re-Identification
• Robust Duality Learning for Unsupervised Visible-Infrared Person Re-Identification
• Multiple Expert Brainstorming for Domain Adaptive Person Re-identification
• Unsupervised Two-Stage Anomaly Detection
• Group-aware Label Transfer for Domain Adaptive Person Re-identification
• Leveraging Ensembles and Self-Supervised Learning for Fully-Unsupervised Person Re-Identification and Text Authorship Attribution
• Self-distillation with Batch Knowledge Ensembling Improves ImageNet Classification
• Joint Noise-Tolerant Learning and Meta Camera Shift Adaptation for Unsupervised Person Re-Identification
• CANU-ReID: A Conditional Adversarial Network for Unsupervised person Re-IDentification
• Self-supervising Fine-grained Region Similarities for Large-scale Image Localization
• Multiple Domain Experts Collaborative Learning: Multi-Source Domain Generalization For Person Re-Identification
• Robust Mutual Learning for Semi-supervised Semantic Segmentation
• ICE: Inter-instance Contrastive Encoding for Unsupervised Person Re-identification
• Dual-Refinement: Joint Label and Feature Refinement for Unsupervised Domain Adaptive Person Re-Identification
• Generalizable Person Re-identification with Relevance-aware Mixture of Experts
• FollowMe: a Robust Person Following Framework Based on Re-Identification and Gestures
• Intra-Inter Camera Similarity for Unsupervised Person Re-Identification
• Joint Generative and Contrastive Learning for Unsupervised Person Re-identification
• ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection
• Semantics-Guided Clustering with Deep Progressive Learning for Semi-Supervised Person Re-identification
• On Universal Black-Box Domain Adaptation
• Camera-aware Proxies for Unsupervised Person Re-Identification
• Re-identification = Retrieval + Verification: Back to Essence and Forward with a New Metric
• Unsupervised Multi-Source Domain Adaptation for Person Re-Identification
• IDM: An Intermediate Domain Module for Domain Adaptive Person Re-ID
• CORE-ReID: Comprehensive Optimization and Refinement through Ensemble fusion in Domain Adaptation for person re-identification
• Improving Unsupervised Domain Adaptive Re-Identification via Source-Guided Selection of Pseudo-Labeling Hyperparameters
• Camera-aware Style Separation and Contrastive Learning for Unsupervised Person Re-identification
• Refining Pseudo Labels with Clustering Consensus over Generations for Unsupervised Object Re-identification
• Improved Mutual Mean-Teaching for Unsupervised Domain Adaptive Re-ID
• TAL: Two-stream Adaptive Learning for Generalizable Person Re-identification
• UnrealPerson: An Adaptive Pipeline towards Costless Person Re-identification
• On Evolving Attention Towards Domain Adaptation
• Unsupervised Pre-training for Person Re-identification
• Rethinking Sampling Strategies for Unsupervised Person Re-identification
• Towards Discriminative Representation Learning for Unsupervised Person Re-identification
• Layerwise Optimization by Gradient Decomposition for Continual Learning
• Discriminability Distillation in Group Representation Learning
• Disentanglement-based Cross-Domain Feature Augmentation for Effective Unsupervised Domain Adaptive Person Re-identification
• Graph Consistency Based Mean-Teaching for Unsupervised Domain Adaptive Person Re-Identification
• Unleashing the Potential of Unsupervised Pre-Training with Intra-Identity Regularization for Person Re-Identification
• COCAS: A Large-Scale Clothes Changing Person Dataset for Re-identification
• Deep Semi-supervised Knowledge Distillation for Overlapping Cervical Cell Instance Segmentation
• Large-Scale Unsupervised Person Re-Identification with Contrastive Learning
• Meta Clustering Learning for Large-scale Unsupervised Person Re-identification
• Unsupervised Domain Generalization for Person Re-identification: A Domain-specific Adaptive Framework
• Unsupervised Domain Adaptive Learning via Synthetic Data for Person Re-identification
• City-Scale Multi-Camera Vehicle Tracking Guided by Crossroad Zones
• Domain Camera Adaptation and Collaborative Multiple Feature Clustering for Unsupervised Person Re-ID
• Multi-organ Segmentation via Co-training Weight-averaged Models from Few-organ Datasets
• Complementary Relation Contrastive Distillation
• Unsupervised Person Re-Identification with Multi-Label Learning Guided Self-Paced Clustering
• Take More Positives: An Empirical Study of Contrastive Learing in Unsupervised Person Re-Identification
• Deep High-Resolution Representation Learning for Cross-Resolution Person Re-identification
• Unsupervised Person Re-identification via Simultaneous Clustering and Consistency Learning
• Cell Detection in Domain Shift Problem Using Pseudo-Cell-Position Heatmap
• An Empirical Study of Vehicle Re-Identification on the AI City Challenge
• Adaptive Pseudo-Label Refinement by Negative Ensemble Learning for Source-Free Unsupervised Domain Adaptation
• Vision Meets Wireless Positioning: Effective Person Re-identification with Recurrent Context Propagation
• Unsupervised Image Segmentation using Mutual Mean-Teaching
• On Leveraging Unlabeled Data for Concurrent Positive-Unlabeled Classification and Robust Generation