The goal of the study is to detect fake images. An autoencoder is trained adversarially to obtain a latent rich representation of real images. The latent representation is then used to reconstruct an image based on the given input image. We propose a comparative study to test if reconstructed features differ strongly for real and fake images leading to enhanced classification performance. We further argue that adversarially training a classifier (DCGAN) generalizes better on unseen data as compared to CNN based architectures.
The goal of this project is to develop an open-source Python package to learn and teach AI, similar to Swirl for R. It aims to build on the work by Alexander Bauer. Course content and the application are currently under development. Stay Tuned!
The goal of this project is to flag fradulent claims using Computer Vision techniques for an automotive insurance company. The idea is to use misalignment in tables and changes in fonts as signals to classify insurance claims.