Adversarial auto‐encoder for unsupervised deep domain adaptation. Issue 14 (31st October 2019)

• Record Type: Journal Article
• Authors: Shao, Rui; Lan, Xiangyuan
• Is Part Of: IET image processing. Volume 13:Issue 14(2019)
• Page Start: 2772
• Page End: 2777
• Publication Date: 2019-10-31
• Subjects: image classification -- pattern classification -- unsupervised learning -- learning (artificial intelligence); unsupervised deep domain adaptation -- unsupervised visual domain adaptation -- target domain -- labelled source samples -- unlabelled target samples -- target domains -- generative adversarial networks -- novel adversarial auto‐encoder -- domain similarity -- reconstruction information -- adversarial domain adaptation; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2018.6687
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : Unsupervised visual domain adaptation aims to train a classifier that works well on a target domain given labelled source samples and unlabelled target samples. The key issue in unsupervised visual domain adaptation is how to do the feature alignment between source and target domains. Inspired by the adversarial learning in generative adversarial networks, this study proposes a novel adversarial auto‐encoder for unsupervised deep domain adaptation. This method incorporates the auto‐encoder with the adversarial learning so that the domain similarity and reconstruction information from the decoder can be exploited to facilitate the adversarial domain adaptation in the encoder. Extensive experiments on various visual recognition tasks show that the proposed method performs favourably against or better than competitive state‐of‐the‐art methods.