Recent advances in anomaly generation approaches alleviate the effect of data insufficiency on task of anomaly localization. While effective, most of them learn multiple large generative models on different datasets and cumbersome anomaly prediction models for different classes. To address the limitations, we propose a novel scalable framework, named AnomalyFactory, that unifies unsupervised anomaly generation and localization with same network architecture. It starts with a BootGenerator that combines structure of a target edge map and appearance of a reference color image with the guidance of a learned heatmap. Then, it proceeds with a FlareGenerator that receives supervision signals from the BootGenerator and reforms the heatmap to indicate anomaly locations in the generated image. Finally, it easily transforms the same network architecture to a BlazeDetector that localizes anomaly pixels with the learned heatmap by converting the anomaly images generated by the FlareGenerator to normal images. By manipulating the target edge maps and combining them with various reference images, AnomalyFactory generates authentic and diversity samples cross domains. Comprehensive experiments carried on 5 datasets, including MVTecAD, VisA, MVTecLOCO, MADSim and RealIAD, demonstrate that our approach is superior to competitors in generation capability and scalability.
Unsupervised training of keypoint-agnostic descriptors for flexible retinal image registration
Grupo VARPA, Instituto de Investigacion Biomédica de A Coruña (INIBIC), Universidade da Coruña(VARPA组,阿维拉生物医学研究所(INIBIC),科鲁纳大学);Departamento de Ciencias de la Computación y Tecnologías de la Información, Universidade da Coruña(计算机科学与信息技术部门,科鲁纳大学)
GAN Inversion for Image Editing via Unsupervised Domain Adaptation
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);University of Virginia(弗吉尼亚大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院)
Parallax Attention for Unsupervised Stereo Correspondence Learning
College of Electronic Science and Technology, Nation University of Defense Technology (NUDT)(国防科技大学电子科学学院);School of Electronics and Communication Engineering, Sun Yat-sen University(中山大学电子与通信工程学院)
Treating Motion as Option with Output Selection for Unsupervised Video Object Segmentation
School of Electrical and Electronic Engineering, Yonsei University(延世大学电气电子工程学院);Department of Software Convergence, Kyung Hee University(韩国庆熙大学软件融合系);EO/IR Systems R&D Lab, LIG Nex1(LIG Nex1 EO/IR系统研发实验室)
Eliminating Warping Shakes for Unsupervised Online Video Stitching
Institute of Information Science, Beijing Jiaotong University(北京交通大学信息科学研究所);Nanyang Technological University(南洋理工大学);Communication University of Zhejiang(浙江传媒学院);University of Electronic Science and Technology of China(电子科技大学)
Mitigating the Influence of Domain Shift in Skin Lesion Classification: A Benchmark Study of Unsupervised Domain Adaptation Methods on Dermoscopic Images
German Aerospace Center (DLR)(德国航空航天中心);German Cancer Research Center (DKFZ)(德国癌症研究中心)
An Explicit Local and Global Representation Disentanglement Framework with Applications in Deep Clustering and Unsupervised Object Detection
Vidyasirimedhi Institute of Science and Technology (VISTEC)(维迪亚西利梅迪科学技术学院);King Mongkut’s University of Technology Thonburi (KMUTT)(国王科技大学吞武里分校);Imperial College London(帝国理工学院)
Unsupervised brain lesion segmentation from MRI using a convolutional autoencoder
University of Iceland(冰岛大学);Landspitali - University Hospital(兰斯皮塔利大学医院);The Icelandic Heart Association(冰岛心脏协会);The Johns Hopkins University(约翰斯·霍普金斯大学)
A coarse-to-fine framework for unsupervised multi-contrast MR image deformable registration with dual consistency constraint
Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(中国科学院深圳先进技术研究院保罗·C·劳特伯生物医学成像研究中心);Pengcheng Laboratory(鹏城实验室);Pazhou Lab(琶洲实验室);Guizhou Provincial People’s Hospital(贵州省人民医院)
Unsupervised Anomaly Instance Segmentation for Baggage Threat Recognition
Khalifa University(哈利法大学);Durham University(杜伦大学);The University of Western Australia(西澳大学);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)
LE-UDA: Label-efficient unsupervised domain adaptation for medical image segmentation
Nanyang Technological University(南洋理工大学);Agency for Science, Technology and Research(新加坡科技研究局);National University of Singapore(新加坡国立大学);Shanghai University(上海大学);University of Science and Technology of China(中国科学技术大学);Chinese Academy of Sciences(中国科学院)
Building RadiologyNET: Unsupervised annotation of a large-scale multimodal medical database
University of Rijeka – Faculty of Engineering(里耶卡大学工程学院);Center for Artificial Intelligence and Cybersecurity(人工智能与网络安全中心);Medical University of Graz(格拉茨医科大学)
UDBE: Unsupervised Diffusion-based Brightness Enhancement in Underwater Images
Centro de Ciências Computacionais(计算机科学中心);Universidade Federal do Rio Grande(南里奥格兰德联邦大学);Instituto de Ciências Exatas e Tecnologia(精确科学与技术研究所);Universidade Federal do Amazonas(亚马逊联邦大学)
Bias in Unsupervised Anomaly Detection in Brain MRI
Technical University of Munich(慕尼黑工业大学);Helmholtz AI(亥姆霍兹人工智能研究所);Helmholtz Center Munich(亥姆霍兹慕尼黑中心);King’s College London(伦敦国王学院);Klinikum Rechts der Isar(伊萨尔河畔临床教学中心);Imperial College London(帝国理工学院)
(Predictable) Performance Bias in Unsupervised Anomaly Detection
Klinikum rechts der Isar der Technischen Universität München(慕尼黑工业大学 rechts der Isar 医院);Technical University of Munich(慕尼黑工业大学);Konrad Zuse School of Excellence in Reliable AI(康拉德·楚泽可靠人工智能卓越学院);Munich Data Science Institute(慕尼黑数据科学研究所);Helmholtz Munich(亥姆霍兹慕尼黑中心);Imperial College London(帝国理工学院)