CycleGAN Facade Generated



Instructor: Casey Rehm



According to the Computer Vision Foundation, image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs.


This work started with generating mass by using non-learning Intelligent agent algorithms and exploring the use of Cycle Consistent Adversarial Networks to create images for the building facades.
The process continued by using CycleGAN to train the images of existing building facades and the renders of the generated masses.