Calculus of Variations and Geometric Measure Theory

G. Batzolis - M. Carioni - C. Etmann - S. Afyouni - Z. Kourtzi - C. B. Schönlieb

CAFLOW: Conditional Autoregressive Flows

created by carioni on 04 Apr 2022
modified on 05 Apr 2022


Submitted Paper

Inserted: 4 apr 2022
Last Updated: 5 apr 2022

Year: 2021

ArXiv: 2106.02531 PDF


We introduce CAFLOW, a new diverse image-to-image translation model that simultaneously leverages the power of auto-regressive modeling and the modeling efficiency of conditional normalizing flows. We transform the conditioning image into a sequence of latent encodings using a multi-scale normalizing flow and repeat the process for the conditioned image. We model the conditional distribution of the latent encodings by modeling the auto-regressive distributions with an efficient multi-scale normalizing flow, where each conditioning factor affects image synthesis at its respective resolution scale. Our proposed framework performs well on a range of image-to-image translation tasks. It outperforms former designs of conditional flows because of its expressive auto-regressive structure.