Computer Vision project carried out as part of the Advanced Master in Artificial Intelligence at Télécom Paris
The main objective of this project is to do domain adaptation. It is to develop a Cycle GAN model to generate SVHN images from MNIST images.
Then, it is to create a CNN that will be trained on the SVHN images generated by the Cycle GAN with the MNIST labs as input. Finally, the model will be tested on the original SVHN dataset.