Semantic Correlation Transfer for Heterogeneous Domain Adaptation

Zhao, Ying Na; Li, Shuang; Zhang, Rui; Liu, Chi Harold; Cao, Weipeng; Wang, Xizhao; Tian, Shu

doi:10.1109/tnnls.2022.3199619

Cited by 5 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, our method SPENet is compared with the existing SSUDA methods [17], [48], [53], [56] and MSUDA methods [11], [45], [63]. All SSUDA methods in Table II are reproduced under two settings.…”

Section: B Comparison With the Existing Methodsmentioning

confidence: 99%

“…In the multisource setting, MSUDA methods, ColorMap-GAN [45] and MADAN [63], adapt to the target domain by changing the style of the data. Their performance in the building extraction task is not satisfactory, only achieving the average IoU of 40.3% and 39.6%.…”

Section: B Comparison With the Existing Methodsmentioning

confidence: 99%

“…The UDA algorithms designed for image classification perform poorly on the semantic segmentation task, because the latter requires dense pixel-level prediction, which increases the difficulty of adaptation. Zhao et al [63] propose a multisource adversarial domain aggregation network (MADAN) for semantic segmentation. The process first translates each source domain style to the target and then aggregates them.…”

Section: Multisource Udamentioning

confidence: 99%

“…(a) ColorMapGAN [45]. (b) MADAN [63]. (c) Multisource domain adaptation with collaborative learning (MSDACL) [11].…”

Section: Introductionmentioning

confidence: 99%

“…They can be classified into two streams: 1) domain alignment-align distributions of multiple-source and target domains at a certain level and 2) self-training-use pseudo-labels of the target domain from different networks to retrain each network. The existing alignment-based methods [45], [63] align multiple domains at the input level by translating the image style, as shown in Fig. 1(a) and (b).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Select, Purify, and Exchange: A Multisource Unsupervised Domain Adaptation Method for Building Extraction

Wang

Zang

Dong

et al. 2024

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

Accurately extracting buildings from aerial images has essential research significance for timely understanding human intervention on the land. The distribution discrepancies between diversified unlabeled remote sensing images (changes in imaging sensor, location, and environment) and labeled historical images significantly degrade the generalization performance of deep learning algorithms. Unsupervised domain adaptation (UDA) algorithms have recently been proposed to eliminate the distribution discrepancies without re-annotating training data for new domains. Nevertheless, due to the limited information provided by a single-source domain, single-source UDA (SSUDA) is not an optimal choice when multitemporal and multiregion remote sensing images are available. We propose a multisource UDA (MSUDA) framework SPENet for building extraction, aiming at selecting, purifying, and exchanging information from multisource domains to better adapt the model to the target domain. Specifically, the framework effectively utilizes richer knowledge by extracting target-relevant information from multiple-source domains, purifying target domain information with low-level features of buildings, and exchanging target domain information in an interactive learning manner. Extensive experiments and ablation studies constructed on 12 city datasets prove the effectiveness of our method against existing state-ofthe-art methods, e.g., our method achieves 59.1% intersection over union (IoU) on Austin and Kitsap −→ Potsdam, which surpasses the target domain supervised method by 2.2%. The code is available at https://github.com/QZangXDU/SPENet.

show abstract

Section: B Comparison With the Existing Methodsmentioning

confidence: 99%

Section: B Comparison With the Existing Methodsmentioning

confidence: 99%

Section: Multisource Udamentioning

confidence: 99%

“…(a) ColorMapGAN [45]. (b) MADAN [63]. (c) Multisource domain adaptation with collaborative learning (MSDACL) [11].…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Select, Purify, and Exchange: A Multisource Unsupervised Domain Adaptation Method for Building Extraction

Wang

Zang

Dong

et al. 2024

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

show abstract

High-order proximity and relation analysis for cross-network heterogeneous node classification

Wu,

et al. 2024

Mach Learn

View full text Add to dashboard Cite

Unsupervised heterogeneous domain adaptation for EEG classification

Wu,

Xie,

et al. 2024

J. Neural Eng.

View full text Add to dashboard Cite

$Objective.$ Domain adaptation has been recognized as a potent solution to the challenge of limited training data for electroencephalography (EEG) classification tasks. Existing studies primarily focus on homogeneous environments, however, the heterogeneous properties of EEG data arising from device diversity cannot be overlooked. This motivates the development of heterogeneous domain adaptation methods that can fully exploit the knowledge from an auxiliary heterogeneous domain for EEG classification. $Approach.$ In this article, we propose a novel model named Informative Representation Fusion (IRF) to tackle the problem of unsupervised heterogeneous domain adaptation in the context of EEG data. In IRF, we consider different perspectives of data, i.e., independent identically distributed (iid) and non-iid, to learn different representations. Specifically, from the non-iid perspective, IRF models high-order correlations among data by hypergraphs and develops hypergraph encoders to obtain data representations of each domain. From the non-iid perspective, by applying multi-layer perceptron networks to the source and target domain data, we achieve another type of representation for both domains. Subsequently, an attention mechanism is used to fuse these two types of representations to yield informative features. To learn transferable representations, the Maximum Mean Discrepancy is utilized to align the distributions of the source and target domains based on the fused features. $Main~results.$ Experimental results on several real-world datasets demonstrate the effectiveness of the proposed model. $Significance.$ This article handles an EEG classification situation where the source and target EEG data lie in different spaces, and what's more, under an unsupervised learning setting. This situation is practical in the real world but barely studied in the literature. The proposed model achieves high classification accuracy, and this study is important for the commercial applications of EEG-based BCIs.

show abstract

Semantic Correlation Transfer for Heterogeneous Domain Adaptation

Cited by 5 publications

References 0 publications

Select, Purify, and Exchange: A Multisource Unsupervised Domain Adaptation Method for Building Extraction

Select, Purify, and Exchange: A Multisource Unsupervised Domain Adaptation Method for Building Extraction

High-order proximity and relation analysis for cross-network heterogeneous node classification

Unsupervised heterogeneous domain adaptation for EEG classification

Contact Info

Product

Resources

About