We extend the work by Golan et al. [1] by implementing the geometric transformations as spatial transformer modules [2]. This code was written during a "Project in Machine Learning" (236757) course at the Technion, Israel Institute of Technology.
[1] Golan, Izhak, and Ran El-Yaniv. "Deep anomaly detection using geometric transformations." Advances in Neural Information Processing Systems (NIPS). 2018.
[2] Jaderberg, Max, Karen Simonyan, and Andrew Zisserman. "Spatial transformer networks." Advances in Neural Information Processing Systems (NIPS). 2015.