Deep visual unsupervised domain adaptation for classification tasks: a survey

Madadi, Yeganeh; Seydi, Vahid; Nasrollahi, Kamal; Hosseini, Reshad; Moeslund, Thomas B.

doi:10.1049/iet-ipr.2020.0087

Cited by 31 publications

(13 citation statements)

References 133 publications

(181 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed annotation transfer pipeline is related to homogenous discrepancy-based domain adaptation methods [31], [53]. Since the feature spaces between the source and target domains are identical and only differ in terms of data distribution, it is similar to homogeneous domain adaptations [53].…”

Section: Discussionmentioning

confidence: 99%

“…Since the feature spaces between the source and target domains are identical and only differ in terms of data distribution, it is similar to homogeneous domain adaptations [53]. Since the reliability measure of the transferred labels can be interpreted as the distance between the source and the target domain, the proposed pipeline is also similar to discrepancy-based domain adaption methods [31]. In contrast to classical domain adaptation methods that utilize labeled data in the source domains to execute new tasks in a target domain [53], the proposed approach solves the same task in the target domain.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

MAAD-Face: A Massively Annotated Attribute Dataset for Face Images

Terhörst

Fährmann

Kolf

et al. 2021

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

Soft-biometrics play an important role in face biometrics and related fields since these might lead to biased performances, threaten the user's privacy, or are valuable for commercial aspects. Current face databases are specifically constructed for the development of face recognition applications. Consequently, these databases contain a large number of face images but lack in the number of attribute annotations and the overall annotation correctness. In this work, we propose a novel annotation-transfer pipeline that allows to accurately transfer attribute annotations from multiple source datasets to a target dataset. The transfer is based on a massive attribute classifier that can accurately state its prediction confidence. Using these prediction confidences, a high correctness of the transferred annotations is ensured. Applying this pipeline to the VGGFace2 database, we propose the MAAD-Face annotation database. It consists of 3.3M faces of over 9k individuals and provides 123.9M attribute annotations of 47 different binary attributes. Consequently, it provides 15 and 137 times more attribute annotations than CelebA and LFW. Our investigation on the annotation quality by three human evaluators demonstrated the superiority of the MAAD-Face annotations over existing databases. Additionally, we make use of the large number of high-quality annotations from MAAD-Face to study the viability of soft-biometrics for recognition, providing insights into which attributes support genuine and imposter decisions. The MAAD-Face annotations dataset is publicly available.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

MAAD-Face: A Massively Annotated Attribute Dataset for Face Images

Terhörst

Fährmann

Kolf

et al. 2021

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

show abstract

“…Whereas, labeled data are available only in the source data in the unsupervised domain adaptation method. Regarding adversarial domain adaptation researches, the unsupervised domain adaptation has been investigated more than two other categories [39].…”

Section: Introductionmentioning

confidence: 99%

A Survey on Adversarial Domain Adaptation

Zonoozi

Seydi

2022

Neural Process Lett

Self Cite

View full text Add to dashboard Cite

Having a lot of labeled data is always a problem in machine learning issues. Even by collecting lots of data hardly, shift in data distribution might emerge because of differences in source and target domains. The shift would make the model to face with problems in test step. Therefore, the necessity of using domain adaptation emerges. There are three techniques in the field of domain adaptation namely discrepancy based, adversarial based and reconstruction based methods. For domain adaptation, adversarial learning approaches showed state-of-the-art performance. Although there are some comprehensive surveys about domain adaptation, we technically focus on adversarial based domain adaptation methods. We examine each proposed method in detail with respect to their structures and objective functions. The common aspect of proposed methods besides domain adaptation is considering the target labels are predicted as accurately as possible. It can be represented by some methods such as metric learning and multi-adversarial discriminators as are used in some of the papers. Also, we address the negative transfer issue for dissimilar distributions and propose the addition of clustering heuristics to the underlying structures for future research.

show abstract

“…In real-world applications, it is typically challenging to obtain sufficient number of annotated training samples. To address this problem, domain adaptation (DA) [1] has been successfully developed to adapt the feature representations learned in the source domain with required label information to the target domain with fewer or even no label information.…”

Section: Introductionmentioning

confidence: 99%

“…Our proposed method is based on domain-invariant features adaptation. This category of methods is obtained through optimizing several measures of domain discrepancy, such as Maximum Mean Discrepancy (MMD) [3,5,6], Low-rank representation [7,8], and Correlation Alignment (CORAL) [9,10,11]. Furthermore, we propose a combination of deep DA methods through the stacking ensemble strategy.…”

Section: Introductionmentioning

confidence: 99%

Stacking Ensemble Learning in Deep Domain Adaptation for Ophthalmic Image Classification

Madadi¹,

Seydi²,

Sun³

et al. 2021

Ophthalmic Medical Image Analysis

Self Cite

View full text Add to dashboard Cite

Domain adaptation is an attractive approach given the availability of a large amount of labeled data with similar properties but different domains. It is effective in image classification tasks where obtaining sufficient label data is challenging. We propose a novel method, named SELDA, for stacking ensemble learning via extending three domain adaptation methods for effectively solving real-world problems. The major assumption is that when base domain adaptation models are combined, we can obtain a more accurate and robust model by exploiting the ability of each of the base models. We extend Maximum Mean Discrepancy (MMD), Low-rank coding, and Correlation Alignment (CORAL) to compute the adaptation loss in three base models. Also, we utilize a two-fully connected layer network as a meta-model to stack the output predictions of these three well-performing domain adaptation models to obtain high accuracy in ophthalmic image classification tasks. The experimental results using Age-Related Eye Disease Study (AREDS) benchmark ophthalmic dataset demonstrate the effectiveness of the proposed model.

show abstract

Deep visual unsupervised domain adaptation for classification tasks: a survey

Cited by 31 publications

References 133 publications

MAAD-Face: A Massively Annotated Attribute Dataset for Face Images

MAAD-Face: A Massively Annotated Attribute Dataset for Face Images

A Survey on Adversarial Domain Adaptation

Stacking Ensemble Learning in Deep Domain Adaptation for Ophthalmic Image Classification

Contact Info

Product

Resources

About