一个异常检测库,包含最先进的算法以及实验管理、超参数优化和边缘推理等功能。
anomaly-localization 开源项目推荐(19 个)
anomaly-localization 话题在 GitHub 上没有填写简介,这里按 Star 数汇总最相关的热门仓库。
热门项目 全部 19
GitHub 共 19 个仓库,已全部展示这是论文“PaDiM:异常检测和定位的补丁分布建模框架”的非官方实现。
Unofficial implementation of EfficientAD https://arxiv.org/abs/2303.14535
This project proposes an end-to-end framework for semi-supervised Anomaly Detection and Segmentation in images based on Deep Learning.
Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoders
[ICLR 2026] 论文《基础视觉编码器是秘密的少量样本异常检测器》的实现
This is an unofficial implementation of Reconstruction by inpainting for visual anomaly detection (RIAD).
GeneralAD
Shape-Guided Dual-Memory Learning for 3D Anomaly Detection [ICML2023]
基于 Anodet 优化的 PaDim
[IJCV 2024] Hard-normal Example-aware Template Mutual Matching for Industrial Anomaly Detection
基于 EfficientNetV2 的 PaDiM
This repository contains code from our comparative study on state of the art unsupervised pathology detection and segmentation methods.
This is an unofficial implementation of ' Anomaly localization by modeling perceptual features'
Anomaly localization using autoencoder models in the feature space of a ResNet
[TF 2.x] PaDiM - unofficial tensorflow implementation of the paper 'a Patch Distribution Modeling Framework for Anomaly Detection and Localization'.
XAVAD: Guided Zero-Shot Anomaly Explanation for Anonymized Crowd Video Monitoring: Context, Categorization, and Spatial Localization.
Group Pattern Recognition project on brain MRI anomaly localization with counterfactual diffusion, MONAI and classifier-free guidance, based on What is Healthy?
Official implementation of PRISM-V, a learning-free approach to visual anomaly detection and localization.
关于这个话题
- 清单取自 GitHub 话题页并按 Star 数排序,每次展示前 19 个项目;原始页面见 https://github.com/topics/anomaly-localization。
- 该话题在 GitHub 上共有 19 个仓库, 本站收录的是其中最热门的 19 个,不是全量列表。
- Star / Fork 每日自动刷新一次,实时数据以 GitHub 为准。
- 当前 19 个项目累计 8233 Star,覆盖 2 种主要语言,数据更新于 2026-11-06。
- 中文介绍来自 GitHub 官方话题说明(github/explore)的机器翻译,可点「查看原文」对照;译文缺失时会先展示原文。