xtarx / unsupervised-anomaly-detection-with-generative-adversarial-networks

Unsupervised Anomaly Detection with Generative Adversarial Networks on MIAS dataset

Anomaly detection using GANs

The goal of this project is be able to detect anomolies using GANs based on Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery

Dataset

we are using MIAS dataset http://peipa.essex.ac.uk/info/mias.html

Architecture

Our code is based on DCGAN from this wonderful repo https://github.com/carpedm20/DCGAN-tensorflow

Results

• TODO

How to run

• TODO

Contributors

• Ahmed Ayad
• M.Tarek Shaban - @XtarX
• Wassim Gabriel - @wassimG

To Do

• Complete Readme.
• Evaluate on 128x128 patches.

License

Apache License 2.0

unsupervised-anomaly-detection-with-generative-adversarial-networks's Issues

The code seemed not to implement the algorithm in the paper, does it?

how to run the code

I found this project to be really useful, feel great appreciated if you could update the readme file about how to run this code. Thanks

how to identify the anomaly areas in unseen image?

May I know how to predict the anomaly areas in the unseen images after training?

Thanks,
Vinod