Chasston's picture
Upload README.md with huggingface_hub
a7dead4 verified
metadata
tags:
  - anomaly-detection
  - computer-vision
  - patchcore
  - defect-detection
  - industrial
datasets:
  - mvtec-ad

DefectVision — PatchCore Anomaly Detection (bottle)

PatchCore implemented from scratch with PyTorch on MVTec AD.

Why PatchCore over YOLOv8?

Model Approach Key Metric
YOLOv8n Supervised mAP50 = 0.09
PatchCore Unsupervised AUROC = 0.9976

YOLOv8 struggles with MVTec AD because defect images are scarce. PatchCore trains on normal images only and generalizes to unseen defects.

Architecture

  • Backbone: WideResNet50 (ImageNet pretrained)
  • Layers: layer2 + layer3 → 1536-dim patch features
  • Memory Bank: coreset 10% of train patches
  • Scoring: max nearest-neighbor distance (k=9)

Metrics — bottle

  • AUROC : 0.9976
  • Best F1: 0.9920

Usage

import torch
from huggingface_hub import hf_hub_download

path = hf_hub_download(repo_id='Chasston/defect-vision-patchcore-bottle', filename='memory_bank.pt')
data = torch.load(path)
memory_bank = data['memory_bank']

Author

Chasston