Print Email Facebook Twitter EAD-GAN Title EAD-GAN: A Generative Adversarial Network for Disentangling Affine Transforms in Images Author Liu, Letao (Nanyang Technological University) Jiang, Xudong (Nanyang Technological University) Saerbeck, Martin (TÜV SÜD PSB) Dauwels, J.H.G. (TU Delft Signal Processing Systems) Date 2022 Abstract This article proposes a generative adversarial network called explicit affine disentangled generative adversarial network (EAD-GAN), which explicitly disentangles affine transform in a self-supervised manner. We propose an affine transform regularizer to force the InfoGAN to have explicit properties of affine transform. To facilitate training an affine transform encoder, we decompose the affine matrix into two separate matrices and infer the explicit transform parameters by the least-squares method. Unlike the existing approaches, representations learned by the proposed EAD-GAN have clear physical meaning, where transforms, such as rotation, horizontal and vertical zooms, skews, and translations, are explicitly learned from training data. Thus, we set different values of each transform parameter individually to generate specifically affine transformed data by the learned network. We show that the proposed EAD-GAN successfully disentangles these attributes on the MNIST, CelebA, and dSprites datasets. EAD-GAN achieves higher disentanglement scores with a large margin compared to the state-of-the-art methods on the dSprites dataset. For example, on the dSprites dataset, EAD-GAN achieves the MIG and DCI score of 0.59 and 0.96 respectively, compared to 0.37 and 0.71, respectively, for the state-of-the-art methods. Subject Affine transformdisentanglementgenerative adversarial network (GAN) To reference this document use: http://resolver.tudelft.nl/uuid:53a21bfb-e8ee-4be2-bc68-4ed14837f3bc DOI https://doi.org/10.1109/TNNLS.2022.3195533 ISSN 2162-237X Source IEEE Transactions on Neural Networks and Learning Systems, 35 (3), 3652-3662 Part of collection Institutional Repository Document type journal article Rights © 2022 Letao Liu, Xudong Jiang, Martin Saerbeck, J.H.G. Dauwels Files PDF EAD-GAN_A_Generative_Adve ... Images.pdf 4.5 MB Close viewer /islandora/object/uuid:53a21bfb-e8ee-4be2-bc68-4ed14837f3bc/datastream/OBJ/view