An unofficial implementation of MemSeg: A semi-supervised method for image surface defect detection using differences and commonalities using PyTorch.
- Clone the repository
git clone https://github.com/ntkhoa95/MemSeg-Defect-Detection
cd MemSeg-Defect-Detection
- Download two datasets
- Describable Textures Dataset (DTD)
https://www.robots.ox.ac.uk/~vgg/data/dtd/
- MVTec Dataset
https://www.mvtec.com/company/research/datasets/mvtec-ad
- Run Before running the training program, please configure the datasets directory in the "./configs" folder
python main.py --object_name capsule
- Inference Mode
voila "inference.ipynb" --port 8866 --Voila.ip 127.0.0.1
Using batch training of 8 MVTec Dataset
| target | AUROC-image | AUROC-pixel | AUPRO-pixel | |
|---|---|---|---|---|
| 0 | leather | 100 | 99.23 | 98.54 |
| 1 | wood | |||
| 2 | carpet | |||
| 3 | capsule | 97.89 | 98.48 | 95.69 |
| 4 | cable | |||
| 5 | metal_nut | |||
| 6 | tile | |||
| 7 | grid | |||
| 8 | bottle | 100 | 98.59 | 95.10 |
| 9 | zipper | |||
| 10 | transistor | |||
| 11 | hazelnut | |||
| 12 | pill | |||
| Average |
@article{DBLP:journals/corr/abs-2205-00908,
author = {Minghui Yang and
Peng Wu and
Jing Liu and
Hui Feng},
title = {MemSeg: {A} semi-supervised method for image surface defect detection
using differences and commonalities},
journal = {CoRR},
volume = {abs/2205.00908},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2205.00908},
doi = {10.48550/arXiv.2205.00908},
eprinttype = {arXiv},
eprint = {2205.00908},
timestamp = {Tue, 03 May 2022 15:52:06 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2205-00908.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}