/inpainting-transformer

This is unofficial implementation of Inpainting Transformer for Anomaly Detection(Intra)

Primary LanguagePython

inpainting-transformer

This is unofficial pytorch implementation of Inpainting Transformer for Anomaly Detection

We support train and test(slow version) for InTra models now. Updates would be coming soon.

Acknowledgement

This code is based on the following repositories: https://github.com/taikiinoue45/RIAD

Prerequisites

  • Python 3
    • torch
    • opencv
    • PIL
    • albumentations

Usage

Train

python main.py --image_dir=../mvtec_anomaly_detection/bottle/ --ckpt=./ckpt/InTra/MVTAD_bottle/

Inference

python main.py --image_dir=../mvtec_anomaly_detection/bottle/ --ckpt=./ckpt/InTra/MVTAD_bottle/ --is_infer

TODO(coming soon!)

Code refactoring

speed up inference

Add pretrained models

update prerequisites

update documents