Unsupervised anomaly detection with generative model, keras implementation
| query image | generated similar image | differece |
|:---:|:---:|:---:|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Unsupervised anomaly detection with DCGAN
First, check directory structure
├── main.py ├── anogan.py ├── weights ├── discriminator.h5 └── generator.h5 └── result └── save the generated images when training
To test this project
$ python main.py
To train a model
$ python main.py --mode train
Then, the training steps(image) will be saved 'result' directory
usage: main.py [-h] [--img_idx IMG_IDX] [--label_idx LABEL_IDX] [--mode MODE]
paper : https://arxiv.org/abs/1703.05921
AnoGAN(code, keras) : https://github.com/yjucho1/anoGAN
AnoGAN(code, tf) : https://github.com/LeeDoYup/AnoGAN