yolov5 deepsort 行人 车辆 跟踪 检测 计数

Official PyTorch code for WACV 2022 paper "CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flows"

天池2019广东工业智造创新大赛 布匹疵点检测 天池水也太深了 季军解决方案

This repo contains code to detect anonmaly by comparing an image with its golden refrences/template matching,originally working on automotive parts,but should be able to generalize well accross other domains.

The project can be used to detect defect on the surface of textiles.

GANomaly: Semi-Supervised Anomaly Detection via Adversarial Training