20.500.12556/RUL-156066 Y-GAN learning dual data representations for anomaly detection in images We propose a novel reconstruction-based model for anomaly detection in image data, called ’Y-GAN’. The model consists of a Y-shaped auto-encoder and represents images in two separate latent spaces. The first captures meaningful image semantics, which are key for representing (normal) training data, whereas the second encodes low-level residual image characteristics. To ensure the dual representations encode mutually exclusive information, a disentanglement procedure is designed around a latent (proxy) classifier. Additionally, a novel representation-consistency mechanism is proposed to prevent information leakage between the latent spaces. The model is trained in a one-class learning setting using only normal training data. Due to the separation of semantically-relevant and residual information, Y-GAN is able to derive informative data representations that allow for efficacious anomaly detection across a diverse set of anomaly detection tasks. The model is evaluated in comprehensive experiments with several recent anomaly detection models using four popular image datasets, i.e., MNIST, FMNIST, CIFAR10, and PlantVillage. Experimental results show that Y-GAN outperforms all tested models by a considerable margin and yields state-of-the-art results. The source code for the model is made publicly available at https://github.com/MIvanovska/Y-GAN. anomaly detection one-class learning deep learning disentangled data representations detekcija anomalij enorazredno učenje globoko učenje true false true Angleški jezik Slovenski jezik Članek v reviji 2024-05-07 10:12:48 2024-05-07 10:12:51 2024-05-27 14:39:13 0000-00-00 00:00:00 2024 0 0 17 str. Vol. 248, art. 123410 15 Avg. 2024 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.93 1873-6793 10.1016/j.eswa.2024.123410 184821507 1-s2.0-S0957417424002756-main.pdf 1-s2.0-S0957417424002756-main.pdf 1 83C971776348434F3D3BF2C4CF9C37DD 9b7c21b1a891de1d34d780f26448ccb7935a05288b5f1113d9a73357bcf6227a d63880e2-0c49-11ef-a791-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=183289 https://www.sciencedirect.com/science/article/pii/S0957417424002756 1 9a10bea4-0c49-11ef-a791-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=183288 Fakulteta za elektrotehniko 0 0 0