Asha Anoosheh

Semester Work
Supervisors: Eirikur Agustsson, Dr. Radu Timofte and Prof. Luc Van Gool

ComboGAN: Unrestricted Scalability for Image Domain Translation

The past year has seen unprecedented leaps in the area of learning-based image translation, namely CycleGAN, by Zhu et al. But experiments so far have been tailored to merely two domains at a time, and scaling them to more would require an quadratic number of models to be trained. And with two-domain models taking days to train on current hardware, the number of domains quickly becomes limited by the time and resources required to process them. In this paper, we propose a multi-component image translation model and training scheme which scales linearly - both in resource consumption and time required - with the number of domains. We made publicly available our paper: and codes: