首页/论文/2011.06288

arXiv 2011.062882020-11-12cs.CVcs.AI

Image Anomaly Detection by Aggregating Deep Pyramidal Representations

University of Udine（乌迪内大学）

Pankaj Mishra, Claudio Piciarelli, Gian Luca Foresti

备注：: Published in First International Conference of Industrial Machine Learning ICPR2020

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

英文摘要相关论文

英文摘要：

Anomaly detection consists in identifying, within a dataset, those samples that significantly differ from the majority of the data, representing the normal class. It has many practical applications, e.g. ranging from defective product detection in industrial systems to medical imaging. This paper focuses on image anomaly detection using a deep neural network with multiple pyramid levels to analyze the image features at different scales. We propose a network based on encoding-decoding scheme, using a standard convolutional autoencoders, trained on normal data only in order to build a model of normality. Anomalies can be detected by the inability of the network to reconstruct its input. Experimental results show a good accuracy on MNIST, FMNIST and the recent MVTec Anomaly Detection dataset

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

Mul-GAD: a semi-supervised graph anomaly detection framework via aggregating multi-view information

Hainan University（海南大学）

Anomaly detection is defined as discovering patterns that do not conform to the expected behavior. Previously, anomaly detection was mostly conducted using traditional shallow learning techniques, but with little improvement. As the emergence of graph neural networks (GNN), graph anomaly detection has been greatly developed. However, recent studies have shown that GNN-based methods encounter challenge, in that no graph anomaly detection algorithm can perform generalization on most datasets. To bridge the tap, we propose a multi-view fusion approach for graph anomaly detection (Mul-GAD). The vi…

2022-12-11cs.AIcs.CV

基于金字塔的Mamba多类无监督异常检测

Pyramid-based Mamba Multi-class Unsupervised Anomaly Detection

University of Udine（乌迪内大学）

针对小异常定位难题，提出基于状态空间模型（SSM）的金字塔扫描策略（PSS），结合预训练编码器与特征级合成异常生成器提取多尺度特征，在MVTec基准上实现AP和AU-PRO各提升1%。

2025-04-04cs.CV

Unsupervised Learning of Multi-level Structures for Anomaly Detection

Shanghai University（上海大学）

The main difficulty in high-dimensional anomaly detection tasks is the lack of anomalous data for training. And simply collecting anomalous data from the real world, common distributions, or the boundary of normal data manifold may face the problem of missing anomaly modes. This paper first introduces a novel method to generate anomalous data by breaking up global structures while preserving local structures of normal data at multiple levels. It can efficiently expose local abnormal structures of various levels. To fully exploit the exposed multi-level abnormal structures, we propose to train …

2021-04-25cs.CV

A Hierarchical Transformation-Discriminating Generative Model for Few Shot Anomaly Detection

Tel Aviv University（特拉维夫大学）；Facebook AI Research（Facebook人工智能研究院）

Anomaly detection, the task of identifying unusual samples in data, often relies on a large set of training samples. In this work, we consider the setting of few-shot anomaly detection in images, where only a few images are given at training. We devise a hierarchical generative model that captures the multi-scale patch distribution of each training image. We further enhance the representation of our model by using image transformations and optimize scale-specific patch-discriminators to distinguish between real and fake patches of the image, as well as between different transformations applied…

2021-04-29cs.CVcs.LG

用于通用工业异常检测的自导航残差 Mamba

Self-Navigated Residual Mamba for Universal Industrial Anomaly Detection

提出SNARM框架，通过测试图像内的自参考学习动态计算残差间和残差内特征，结合多头的Mamba模块聚焦异常区域，在多个工业异常检测基准上达到SOTA性能。

2025-08-10cs.CV

Q-Former Autoencoder：用于医学异常检测的现代框架

Q-Former Autoencoder: A Modern Framework for Medical Anomaly Detection

University of Bologna（博洛尼亚大学）；Middle East Technical University（中东技术大学）；Helmholtz Munich（慕尼黑海德堡研究所）

本文提出 Q-Former Autoencoder 无监督医学异常检测框架，利用冻结视觉基础模型、Q-Former 瓶颈与 Masked Autoencoder 感知损失，在多个医学基准上达到最先进性能。

2025-07-24cs.CV

更多相关论文收起更多论文

探索用于多类别无监督异常检测的普通 ViT 重建

Exploring Plain ViT Reconstruction for Multi-class Unsupervised Anomaly Detection

Youtu Lab, Tencent（腾讯优图实验室）；Zhejiang University（浙江大学）；Nanyang Technological University（南洋理工大学）；University of California, Merced（加州大学默塞德分校）

本文面向多类别无监督异常检测，提出仅用普通 ViT 特征的重建模型 ViTAD，在多个数据集上以更低训练成本取得最先进性能。

2024-08-11cs.CV

面向视频异常检测的网络系统：教程与综述

Networking Systems for Video Anomaly Detection: A Tutorial and Survey

Fudan University（复旦大学）；The University of British Columbia（不列颠哥伦比亚大学）；Duke Kunshan University（昆山杜克大学）；Soochow University（苏州大学）；University of Toronto（多伦多大学）；The University of Hong Kong（香港大学）；Massachusetts Institute of Technology（麻省理工学院）；University of Ottawa（渥太华大学）

本文全面综述了视频异常检测（VAD）网络系统（NSVAD），阐述了深度学习驱动的VAD路径的基础假设、学习框架与适用场景，并展望了AI与计算技术整合的未来趋势与机遇。

2025-04-03cs.AIcs.CVcs.CY

看得更多，检测得更少？抑制多视图异常检测中的信息泄漏

See More, Detect Less? Taming Information Leakage in Multi-View Anomaly Detection

National Taiwan University（台湾大学）；Mitsubishi Electric Research Laboratories (MERL)（三菱电机研究实验室）；National Taiwan Normal University（台湾师范大学）；VinUniversity；Microsoft Taiwan Corporation（微软台湾公司）；National Taiwan University of Science and Technology（台湾科技大学）

该研究针对多视图异常检测中的跨视图信息泄漏问题，提出GLAD框架，结合视觉基础模型特征与MMA、OGA模块，在Real-IAD等数据集上优于现有最优方法，证实信息限制的重要性。

2026-08-27cs.CVcs.MM

在数字病理学中利用金字塔图卷积网络整合多尺度拓扑结构

Integrating multiscale topology in digital pathology with pyramidal graph convolutional networks

F. Hoffmann-La Roche AG（罗氏公司）

提出多尺度图卷积网络MS-GCN，通过整合WSI多放大倍数信息同时建模长程结构与细胞细节，克服过度平滑问题，性能与可解释性优于单倍数方法。

2024-03-22cs.CVeess.IV

cOOpD：利用对比表示将胸部 CT 扫描中的 COPD 分类重构为异常检测

cOOpD: Reformulating COPD classification on chest CT scans as anomaly detection using contrastive representations

German Cancer Research Center（德国癌症研究中心）；Heidelberg University（海德堡大学）；Translational Lung Research Center Heidelberg（海德堡转化肺研究中心）；German Center for Lung Research（德国肺研究中心）；Helmholtz Imaging（亥姆霍兹影像中心）

本文提出 cOOpD，将 COPD 胸部 CT 二分类重构为基于对比表示的 OOD 异常检测，在两个公开数据集上显著提升 AUROC，并提供可解释的早期病变定位。

2023-07-14cs.CVeess.IV

基于视觉基础模型的对抗鲁棒小样本异常检测

Adversarially Robust Few-Shot Anomaly Detection with Vision Foundation Models

Rochester Institute of Technology（罗切斯特理工学院）

该研究针对视觉基础模型的小样本异常检测场景，提出无需训练骨干网络的对抗鲁棒方法，经实验在多数据集上提升了对抗检测性能，且保持较高干净准确率。

2026-08-11cs.CV

使用聚合规范扩散的无监督异常检测

Unsupervised Anomaly Detection using Aggregated Normative Diffusion

University of Tübingen（图宾根大学）

本文提出无监督异常检测方法ANDi，通过在金字塔高斯噪声上训练的DDPM中聚合预测去噪步骤与真实反向转移的差异，在多个脑MRI数据集上显著提升了对多样化异常（尤其是MS病变）的检测性能。

2023-12-04cs.CVcs.LGeess.IV

AnomalyClaw：通过工具引导的反驳实现通用视觉异常检测代理

AnomalyClaw: A Universal Visual Anomaly Detection Agent via Tool-Grounded Refutation

Department of Computer Science and Engineering, Southern University of Science and Technology (SUSTech), Shenzhen, China（南方科技大学计算机科学与工程系，深圳，中国）；School of EEE, Nanyang Technological University (NTU), Singapore（南洋理工大学电子工程学院，新加坡）；CFAR, Agency for Science, Technology and Research (A*STAR), Singapore（科技研究局（A*STAR）的CFAR，新加坡）

本文提出AnomalyClaw，一种无需训练的视觉异常检测代理，通过多轮反驳过程提升异常判断。在CrossDomainVAD-12基准上，AnomalyClaw在多个模型上均取得显著提升，且引入自演化扩展提升模型性能。

2026-05-11cs.AIcs.CV

Dense 3D Point Cloud Reconstruction Using a Deep Pyramid Network

Video Analytics Lab, CDS, Indian Institute of Science, Bangalore（印度科学学院计算机与数据科学系视频分析实验室）

Reconstructing a high-resolution 3D model of an object is a challenging task in computer vision. Designing scalable and light-weight architectures is crucial while addressing this problem. Existing point-cloud based reconstruction approaches directly predict the entire point cloud in a single stage. Although this technique can handle low-resolution point clouds, it is not a viable solution for generating dense, high-resolution outputs. In this work, we introduce DensePCR, a deep pyramidal network for point cloud reconstruction that hierarchically predicts point clouds of increasing resolution.…

2019-01-25cs.CV

无需训练的3D脑部MRI中零样本异常检测

Training-Free Zero-Shot Anomaly Detection in 3D Brain MRI with 2D Foundation Models

Department of Mathematics, Chungnam National University（Chungnam国立大学数学系）

本文提出无需训练的3D脑部MRI零样本异常检测框架，通过聚合多轴切片的2D基础模型生成局部体积标记，恢复立方空间上下文并直接整合距离基异常检测流程，提供高效的3D表示。

2026-02-17cs.CVstat.ML

Flow Mismatching: 通过流匹配模型中的速度差异进行无监督异常检测

Flow Mismatching: Unsupervised Anomaly Detection via Velocity Discrepancies in Flow Matching Models

Arizona State University（亚利桑那州立大学）；Georgia Institute of Technology（佐治亚理工学院）

提出Flow Mismatching方法，利用正常数据训练的流匹配模型在测试图像路径上的预测速度与几何速度的差异进行无监督异常检测，无需重建、测试时优化或额外校准。

2026-05-21cs.CV