Bi-Directional Generation for Unsupervised Domain Adaptation

Yang, Guanglei; Xia, Haifeng; Ding, Mingli; Ding, Zhengming

doi:10.1609/aaai.v34i04.6137

Cited by 64 publications

(21 citation statements)

References 26 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• The first group includes 13 state-of-the-art UDA methods requiring access to the source data, i.e., DANN [46], CDAN [15], CAT [47], BSP [48], SAFN [44], SWD [49], ADR [50], TN [51], IA [52], BNM [53], BDG [54], MCC [55] and SRDC [36]. • The second group comprises four state-of-the-art methods for UDA without access to the source data.…”

Section: Baseline Methodsmentioning

confidence: 99%

Nearest Neighborhood-Based Deep Clustering for Source Data-absent Unsupervised Domain Adaptation

Tang,

Yang,

et al. 2021

Preprint

View full text Add to dashboard Cite

In the classic setting of unsupervised domain adaptation (UDA), the labeled source data are available in the training phase. However, in many real-world scenarios, owing to some reasons such as privacy protection and information security, the source data is inaccessible, and only a model trained on the source domain is available. This paper proposes a novel deep clustering method for this challenging task. Aiming at the dynamical clustering at feature-level, we introduce extra constraints hidden in the geometric structure between data to assist the process. Concretely, we propose a geometry-based constraint, named semantic consistency on the nearest neighborhood (SCNNH), and use it to encourage robust clustering. To reach this goal, we construct the nearest neighborhood for every target data and take it as the fundamental clustering unit by building our objective on the geometry. Also, we develop a more SCNNH-compliant structure with an additional semantic credibility constraint, named semantic hyper-nearest neighborhood (SHNNH). After that, we extend our method to this new geometry. Extensive experiments on three challenging UDA datasets indicate that our

show abstract

Section: Baseline Methodsmentioning

confidence: 99%

Nearest Neighborhood-Based Deep Clustering for Source Data-absent Unsupervised Domain Adaptation

Tang,

Yang,

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The theses method does not optimize the segmentation model in the target-to-source directions. Yang et al [56] proposed a bi-directional generation network that trained a simple framework for image translation and classification from source to target and from target to source. Jiang et al [57] proposed a bidirectional adversarial training method which performs adversarial trainings with generating adversarial examples from source to target and back.…”

Section: Bidirectional Learningmentioning

confidence: 99%

BiFDANet: Unsupervised Bidirectional Domain Adaptation for Semantic Segmentation of Remote Sensing Images

et al. 2022

View full text Add to dashboard Cite

When segmenting massive amounts of remote sensing images collected from different satellites or geographic locations (cities), the pre-trained deep learning models cannot always output satisfactory predictions. To deal with this issue, domain adaptation has been widely utilized to enhance the generalization abilities of the segmentation models. Most of the existing domain adaptation methods, which based on image-to-image translation, firstly transfer the source images to the pseudo-target images, adapt the classifier from the source domain to the target domain. However, these unidirectional methods suffer from the following two limitations: (1) they do not consider the inverse procedure and they cannot fully take advantage of the information from the other domain, which is also beneficial, as confirmed by our experiments; (2) these methods may fail in the cases where transferring the source images to the pseudo-target images is difficult. In this paper, in order to solve these problems, we propose a novel framework BiFDANet for unsupervised bidirectional domain adaptation in the semantic segmentation of remote sensing images. It optimizes the segmentation models in two opposite directions. In the source-to-target direction, BiFDANet learns to transfer the source images to the pseudo-target images and adapts the classifier to the target domain. In the opposite direction, BiFDANet transfers the target images to the pseudo-source images and optimizes the source classifier. At test stage, we make the best of the source classifier and the target classifier, which complement each other with a simple linear combination method, further improving the performance of our BiFDANet. Furthermore, we propose a new bidirectional semantic consistency loss for our BiFDANet to maintain the semantic consistency during the bidirectional image-to-image translation process. The experiments on two datasets including satellite images and aerial images demonstrate the superiority of our method against existing unidirectional methods.

show abstract

“…(2) UDA methods: ADDA (Tzeng et al 2017), ADR , MCD (Saito, Watanabe et al 2018), CDAN (Long, Cao et al 2018), CyDADA (Hoffman, Tzeng et al 2018), SAFN (Xu, Li et al 2019), SWD (Lee et al 2019a), TPN (Pan, Yao et al 2019), CAT (Deng, Luo, and Zhu 2019), MDD (Zhang, Liu et al 2019), SWD (Lee et al 2019b), BDG (Yang, Xia et al 2020), PAL (Hu, Liang et al 2020), MCC (Jin, Wang et al 2020), BNM (Cui, Wang et al 2020), CoSCA (Dai, Cheng et al 2020) and SRDC (Tang, Chen, and Jia 2020); (3) SFDA methods: PrDA (Kim, Cho et al 2020), SHOT (Liang, Hu et al 2020), MA (Li, Jiao et al 2020), BAIT (Yang, Wang et al 2020) and CPGA (Qiu et al 2021).…”

Section: Baselinesmentioning

confidence: 99%

Generation, augmentation, and alignment: A pseudo-source domain based method for source-free domain adaptation

Du¹,

Yang²,

Chen³

et al. 2021

Preprint

View full text Add to dashboard Cite

Conventional unsupervised domain adaptation (UDA) methods need to access both labeled source samples and unlabeled target samples simultaneously to train the model. While in some scenarios, the source samples are not available for the target domain due to data privacy and safety. To overcome this challenge, recently, source-free domain adaptation (SFDA) has attracted the attention of researchers, where both a trained source model and unlabeled target samples are given. Existing SFDA methods either adopt a pseudo-label based strategy or generate more samples. However, these methods do not explicitly reduce the distribution shift across domains, which is the key to a good adaptation. Although there are no source samples available, fortunately, we find that some target samples are very similar to the source domain and can be used to approximate the source domain. This approximated domain is denoted as the pseudo-source domain. In this paper, inspired by this observation, we propose a novel method based on the pseudo-source domain. The proposed method firstly generates and augments the pseudo-source domain, and then employs distribution alignment with four novel losses based on pseudo-label based strategy. Among them, a domain adversarial loss is introduced between the pseudo-source domain the remaining target domain to reduce the distribution shift. The results on three real-world datasets verify the effectiveness of the proposed method.

show abstract

Bi-Directional Generation for Unsupervised Domain Adaptation

Cited by 64 publications

References 26 publications

Nearest Neighborhood-Based Deep Clustering for Source Data-absent Unsupervised Domain Adaptation

Nearest Neighborhood-Based Deep Clustering for Source Data-absent Unsupervised Domain Adaptation

BiFDANet: Unsupervised Bidirectional Domain Adaptation for Semantic Segmentation of Remote Sensing Images

Generation, augmentation, and alignment: A pseudo-source domain based method for source-free domain adaptation

Contact Info

Product

Resources

About