Awsome Domain-Adaptation

2018-08-06 19:27:54

 

This blog is copied from:  

 

This repo is a collection of AWESOME things about domian adaptation,including papers,code etc.Feel free to star and fork.

Contents

Papers

Overview

• Deep Visual Domain Adaptation: A Survey 
• Domain Adaptation for Visual Applications: A Comprehensive Survey 

Theory

• Analysis of Representations for Domain Adaptation 
• A theory of learning from different domains 
• Learning Bounds for Domain Adaptation 

Unsupervised DA

Adversarial Methods

• M-ADDA: Unsupervised Domain Adaptation with Deep Metric Learning  
• Augmented Cyclic Adversarial Learning for Domain Adaptation 
• Factorized Adversarial Networks for Unsupervised Domain Adaptation 
• DiDA: Disentangled Synthesis for Domain Adaptation 
• Unsupervised Domain Adaptation with Adversarial Residual Transform Networks 
• Simple Domain Adaptation with Class Prediction Uncertainty Alignment 
• Causal Generative Domain Adaptation Networks 
• Conditional Adversarial Domain Adaptation 
• Deep Adversarial Attention Alignment for Unsupervised Domain Adaptation: the Benefit of Target Expectation Maximization 
• Learning Semantic Representations for Unsupervised Domain Adaptation  
• CyCADA: Cycle-Consistent Adversarial Domain Adaptation  
• From source to target and back: Symmetric Bi-Directional Adaptive GAN   
• Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation 
• Maximum Classifier Discrepancy for Unsupervised Domain Adaptation  
• Domain Generalization with Adversarial Feature Learning 
• Adversarial Feature Augmentation for Unsupervised Domain Adaptation  
• Duplex Generative Adversarial Network for Unsupervised Domain Adaptation  
• Generate To Adapt: Aligning Domains using Generative Adversarial Networks  
• Image to Image Translation for Domain Adaptation 
• Unsupervised Domain Adaptation with Similarity Learning 
• Conditional Generative Adversarial Network for Structured Domain Adaptation 
• Collaborative and Adversarial Network for Unsupervised Domain Adaptation  
• Re-Weighted Adversarial Adaptation Network for Unsupervised Domain Adaptation 
• Multi-Adversarial Domain Adaptation  
• Wasserstein Distance Guided Representation Learning for Domain Adaptation  
• Incremental Adversarial Domain Adaptation for Continually Changing Environments 
• A DIRT-T Approach to Unsupervised Domain Adaptation  
• Label Efficient Learning of Transferable Representations acrosss Domains and Tasks  
• Addressing Appearance Change in Outdoor Robotics with Adversarial Domain Adaptation 
• Adversarial Discriminative Domain Adaptation   
• Unsupervised Pixel–Level Domain Adaptation with Generative Adversarial Networks  
• Domain Separation Networks 
• Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation 
• Domain-Adversarial Training of Neural Networks 
• Unsupervised Domain Adaptation by Backpropagation    

Network Methods

• Boosting Domain Adaptation by Discovering Latent Domains 
• Residual Parameter Transfer for Deep Domain Adaptation 
• Deep Asymmetric Transfer Network for Unbalanced Domain Adaptation 
• Deep CORAL: Correlation Alignment for Deep Domain Adaptation 
• Deep Domain Confusion: Maximizing for Domain Invariance 

Optimal Transport

• DeepJDOT: Deep Joint distribution optimal transport for unsupervised domain adaptation 
• Joint Distribution Optimal Transportation for Domain Adaptation   

Incremental Methods

• Incremental Adversarial Domain Adaptation for Continually Changing Environments 
• Continuous Manifold based Adaptation for Evolving Visual Domains 

Other Methods

• Unsupervised Domain Adaptation with Distribution Matching Machines 
• Self-Ensembling for Visual Domain Adaptation 
• Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation 
• Aligning Infinite-Dimensional Covariance Matrices in Reproducing Kernel Hilbert Spaces for Domain Adaptation 
• Associative Domain Adaptation  
• Learning Transferrable Representations for Unsupervised Domain Adaptation 

Zero-shot DA

• Zero-Shot Deep Domain Adaptation 

Few-shot DA

Image-to-Image Translation

• JointGAN: Multi-Domain Joint Distribution Learning with Generative Adversarial Nets  
• Multimodal Unsupervised Image-to-Image Translation  
• StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation 
• Conditional Image-to-Image Translation 
• Toward Multimodal Image-to-Image Translation   
• Unsupervised Image-to-Image Translation Networks  
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks 
• Image-to-Image Translation with Conditional Adversarial Nets   
• Learning to Discover Cross-Domain Relations with Generative Adversarial Networks  
• Unsupervised Cross-Domain Image Generation  
• Coupled Generative Adversarial Networks  

Open Set DA

• Learning Factorized Representations for Open-set Domain Adaptation 
• Open Set Domain Adaptation by Backpropagation 
• Open Set Domain Adaptation 

Partial DA

• Partial Adversarial Domain Adaptation  
• Importance Weighted Adversarial Nets for Partial Domain Adaptation 
• Partial Transfer Learning with Selective Adversarial Networks  

Multi source DA

• Deep Cocktail Network: Multi-source Unsupervised Domain Adaptation with Category Shift 

Applications

Object Detection

• Cross-Domain Weakly-Supervised Object Detection Through Progressive Domain Adaptation 
• Domain Adaptive Faster R-CNN for Object Detection in the Wild 

Semantic Segmentation

• Learning From Synthetic Data: Addressing Domain Shift for Semantic Segmentation 
• Curriculum Domain Adaptation for Semantic Segmentation of Urban Scenes 

Person Re-identification

• Person Transfer GAN to Bridge Domain Gap for Person Re-Identification 
• Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification 

Others

• Real-Time Monocular Depth Estimation using Synthetic Data with Domain Adaptation via Image Style Transfer 

Benchmarks

• Syn2Real: A New Benchmark forSynthetic-to-Real Visual Domain Adaptation