patchcore

Pipeline training and inference Anomalib models UI in Anomaly Detection

PatchCore method for Industrial Anomaly Detection + CLIP

Production visual anomaly detection from normal images with PaDiM, ultra-light PatchCore, Production Autopilot, ONNX, OpenVINO, and TensorRT INT8 export.

An unofficial implementation using Pytorch for "Anomaly Clustering: Grouping Images into Coherent Clusters of Anomaly Types". Improve the algorithm with DINO pretrained ViT. Implement algorithms based on PatchCore.

Comprehensive benchmark for anomaly detection models on MVTec AD dataset using Anomalib. Includes PatchCore, EfficientAD, FastFlow, STFPM, PaDiM with Gradio demo.

Anomaly detection for visual defect inspection: trained on normal images only, with localisation. PyTorch + MVTec AD.

PatchCore + YOLO screw anomaly detection demo with Gradio UI (MVTec AD)

Multi-view anomaly scoring models from scarce, unbalanced, multi-class data

End-to-end automated visual inspection: PatchCore anomaly detection + FastAPI + Next.js dashboard + ABB GoFa cobot (RobotStudio)

Industrial visual anomaly detection using PatchCore with FastAPI, Streamlit and Docker deployment.

Training-free deep learning for unsupervised industrial defect detection & localization on MVTec AD, PatchCore & PaDiM, with image and pixel level AUROC and a one-click free-Colab demo.

AI-powered industrial surface defect detection using PatchCore-inspired anomaly detection, ResNet50 feature extraction, heatmap visualization, Google Gemini AI, PDF inspection reports, and Streamlit.

📸🔥Thermal Camera Diagnosis App Prototype was my bachelor thesis project at THU, focused on enhancing battery recycling processes using thermal imaging anomaly detection.

Edge anomaly detection for video streams with multi-backend inference benchmarking (TensorRT, ONNX, PyTorch)

Unsupervised maize disease detection with cross-domain evaluation on real-field PlantDoc imagery and a controlled background-sensitivity experiment. PatchCore generalizes from lab to field (AUROC 0.9995); reconstruction-based methods are 4× inflated by domain shift.

兼容 Anomalib 生态的通用异常检测推理库：一张图进去，缺陷 mask + 分数 + OK/NG 出来，一个 DLL + 一个头文件，零 OpenCV 依赖。

Unsupervised anomaly detection pipeline for real-time welding quality control with IATF 16949 audit trail compliance.

Real-time manufacturing defect detection using unsupervised PatchCore — FastAPI + Streamlit + OpenVINO

Modular 2D industrial inspection platform with a swappable method interface — two anomaly-detection methods (autoencoder + PatchCore) built from scratch, served via FastAPI + React, Dockerized. Benchmarked on MVTec AD (mean image AUROC 0.72 → 0.96).

Unsupervised visual defect detection on industrial parts: a reconstruction autoencoder vs an embedding memory bank (PatchCore-lite) on MVTec AD, trained only on good images, on an AMD GPU via ROCm.