本文主要是介绍GAN domian adaptation，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！
GAN 发展：domain 变多
DC GAN
cycle GAN
Combo GAN
Star GAN
DA：
基于分布对齐， 基于 对抗等等
一般：one source one target
multi-source domain adaptation
multi-target domain adaptation
深度域自适应目标检测（DDAOD） - 知乎
【深度域自适应】一、DANN与梯度反转层（GRL）详解 - 知乎
《迁移学习》: 领域自适应(Domain Adaptation)的理论分析 - 知乎
IBN-Net: 提升模型的域自适应性 - 知乎
Deep Domain Adaptation论文集(一)：基于label迁移知识 - 知乎
Deep Domain Adaptation论文集(二)：基于统计差异 - 知乎
Deep Domain Adaptation论文集(三)：基于深度网络结构差异&几何差异 - 知乎
Deep Domain Adaptation论文集(四)：基于生成对抗网络GAN - 知乎
Deep Domain Adaptation论文集(五)：基于数据重构的迁移方法 - 知乎
Deep Domain Adaptation论文集(六)：源域与目标域特征空间不一致的处理方法 - 知乎
《A Survey of Unsupervised Deep Domain Adaptation》 - 知乎
Domain Generalization Notes - 知乎
域适应方法顶刊论文列表 近两年（2020-2021） - 知乎
领域迁移(domain adaptation)经典理论推导 - 知乎
理解深度网络的迁移性 - 知乎
迁移学习之多类域适应（Multi-Class Domain Adaptation）理论和算法总结 - 知乎
Domain-Adversarial Training of Neural Networks
paper：
源域样本不平衡的 DA
https://openaccess.thecvf.com/content/WACV2021/papers/Wang_Towards_Fair_Cross-Domain_Adaptation_via_Generative_Learning_WACV_2021_paper.pdf
cross task cross domain DA
https://openaccess.thecvf.com/content/WACV2021/papers/Chavhan_ADA-ATDT_An_Adversarial_Approach_for_Cross-Domain_and_Cross-Task_Knowledge_Transfer_WACV_2021_paper.pdf
meta-domain-invariant representations
https://openaccess.thecvf.com/content/WACV2021/papers/Sharma_Unsupervised_Meta-Domain_Adaptation_for_Fashion_Retrieval_WACV_2021_paper.pdf
Domain-Adaptive Few-Shot Learning
https://openaccess.thecvf.com/content/WACV2021/papers/Zhao_Domain-Adaptive_Few-Shot_Learning_WACV_2021_paper.pdf
对抗增强学习 进行 域自适应
https://openaccess.thecvf.com/content/WACV2021/papers/Zhang_Adversarial_Reinforcement_Learning_for_Unsupervised_Domain_Adaptation_WACV_2021_paper.pdf
无需源域数据 的 特征对齐
https://openaccess.thecvf.com/content/WACV2021/papers/Yeh_SoFA_Source-Data-Free_Feature_Alignment_for_Unsupervised_Domain_Adaptation_WACV_2021_paper.pdf
MTDA ： 通过知识蒸馏 用于语义分割
https://openaccess.thecvf.com/content/CVPR2021/papers/Isobe_Multi-Target_Domain_Adaptation_With_Collaborative_Consistency_Learning_CVPR_2021_paper.pdf
https://openaccess.thecvf.com/content/WACV2021/papers/Toldo_Unsupervised_Domain_Adaptation_in_Semantic_Segmentation_via_Orthogonal_and_Clustered_WACV_2021_paper.pdf
[cvpr2021] Multi-Source Domain Adaptation with Collaborative Learning for Semantic Segmentation
与 multi target 同一团队
MTDA 声影分类
Unsupervised Multi-Target Domain Adaptation for Acoustic Scene Classification
https://arxiv.org/pdf/2105.10340.pdf
MTDA 多个无标签目标域 自适应
https://openaccess.thecvf.com/content/ICCV2021/papers/Saporta_Multi-Target_Adversarial_Frameworks_for_Domain_Adaptation_in_Semantic_Segmentation_ICCV_2021_paper.pdf https://openaccess.thecvf.com/content/WACV2021/papers/Zhao_Domain-Adaptive_Few-Shot_Learning_WACV_2021_paper.pdf
https://openaccess.thecvf.com/content/WACV2021/papers/Le_Thanh_Nguyen-Meidine_Unsupervised_Multi-Target_Domain_Adaptation_Through_Knowledge_Distillation_WACV_2021_paper.pdf
https://openaccess.thecvf.com/content/CVPR2021/papers/Isobe_Multi-Target_Domain_Adaptation_With_Collaborative_Consistency_Learning_CVPR_2021_paper.pdf
Unsupervised multitarget domain adaptation: An information theoretic approach. IEEE Transactions on Image Processing, 2020.
Multi-target unsupervised domain adaptation without exactly shared categories. CoRR, abs/1809.00852, 2018.
Multi-target Unsupervised Domain Adaptation without Exactly Shared Categories
Heterogeneous Graph Attention Network for Unsupervised Multiple-Target Domain Adaptation
Attention Guided Multiple Source and Target Domain Adaptation
Multi-Target Domain Adaptation via Unsupervised Domain Classification for Weather Invariant Object Detection
domain adaptation开发域不变的数据结构，采用迁移学习，减轻不同域数据分布的区别。训练一个style transfer模型，从各个混合（天气）域提取style features。k-means聚类将图像分成不同天气的类，训练的source-target domain style transfer models可以产生目标域的标注图像，最后是训练一个weather-invariant object detector。
Domain Adversarial Neural Networks for Large-Scale Land Cover Classification
In this work, we propose an unsupervised DA technique called domain adversarial neural networks (DANNs), composed of a feature extractor, a class predictor, and domain classifier blocks, for large-scale land cover classification. Contrary to the traditional methods that perform representation and classifier learning in separate stages, DANNs combine them into a single stage, thereby learning a new representation of the input data that is both domain-invariant and discriminative. Once trained, the classifier of a DANN can be used to predict both source and target domain labels. Additionally, we also modify the domain classifier of a DANN to evaluate its suitability for multi-target domain adaptation problems.
特征分解 feature disentanglement
Reid
http://robotics.pkusz.edu.cn/static/papers/IJCAI2021-xuwanlu.pdff
Informative Feature Disentanglement for Unsupervised Domain Adaptation
Informative Feature Disentanglement for Unsupervised Domain Adaptation | IEEE Journals & Magazine | IEEE Xplore
风格迁移
Multi-mapping Image-to-Image Translation via Learning Disentanglement
https://proceedings.neurips.cc/paper/2019/file/5a142a55461d5fef016acfb927fee0bd-Paper.pdf
A Unified Feature Disentangler for Multi-Domain Image Translation and Manipulation
https://arxiv.org/pdf/1809.01361.pdf
starGAN
https://arxiv.org/pdf/1912.01865.pdf
Cross-Modal Prototypes for Cross-Domain Visual-Language Retrieval
【牛津大学&将最大化互信息引入到无监督域适应(UDA)任务】Structure-Aware Feature Fusion for Unsupervised Domain Adaptation - 知乎
AAAI 2021
论文：Mind-the-Gap! Unsupervised Domain Adaptation for Text-Video Retrieval
论文地址：Mind-the-Gap! Unsupervised Domain Adaptation for Text-Video Retrieval
笔记链接：李加贝：[跨域跨模态检索-1]Mind-the-Gap! Unsupervised Domain Adaptation for Text-Video Retrieval
CVPR2021
论文：Adaptive Cross-Modal Prototypes for Cross-Domain Visual-Language Retrieval
论文地址：https://openaccess.thecvf.com/c
笔记链接：李加贝：[跨域跨模态检索-2]Adaptive Cross-Modal Prototypes for Cross-Domain Visual-Language Retrieval
泛检索领域
A Decade Survey of Content Based Image Retrieval using Deep Learning | IEEE Journals & Magazine | IEEE Xplore
这篇关于GAN domian adaptation的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！
