Invertible Autoencoder for domain adaptation

by   Yunfei Teng, et al.

The unsupervised image-to-image translation aims at finding a mapping between the source (A) and target (B) image domains, where in many applications aligned image pairs are not available at training. This is an ill-posed learning problem since it requires inferring the joint probability distribution from marginals. Joint learning of coupled mappings F_AB: A → B and F_BA: B → A is commonly used by the state-of-the-art methods, like CycleGAN [Zhu et al., 2017], to learn this translation by introducing cycle consistency requirement to the learning problem, i.e. F_AB(F_BA(B)) ≈ B and F_BA(F_AB(A)) ≈ A. Cycle consistency enforces the preservation of the mutual information between input and translated images. However, it does not explicitly enforce F_BA to be an inverse operation to F_AB. We propose a new deep architecture that we call invertible autoencoder (InvAuto) to explicitly enforce this relation. This is done by forcing an encoder to be an inverted version of the decoder, where corresponding layers perform opposite mappings and share parameters. The mappings are constrained to be orthonormal. The resulting architecture leads to the reduction of the number of trainable parameters (up to 2 times). We present image translation results on benchmark data sets and demonstrate state-of-the art performance of our approach. Finally, we test the proposed domain adaptation method on the task of road video conversion. We demonstrate that the videos converted with InvAuto have high quality and show that the NVIDIA neural-network-based end-to-end learning system for autonomous driving, known as PilotNet, trained on real road videos performs well when tested on the converted ones.


page 8

page 11

page 12

page 13

page 14

page 15

page 18

page 19


MCMI: Multi-Cycle Image Translation with Mutual Information Constraints

We present a mutual information-based framework for unsupervised image-t...

Domain Adaptation for Image Dehazing

Image dehazing using learning-based methods has achieved state-of-the-ar...

Similarity-preserving Image-image Domain Adaptation for Person Re-identification

This article studies the domain adaptation problem in person re-identifi...

Motion Blur removal via Coupled Autoencoder

In this paper a joint optimization technique has been proposed for coupl...

Asymmetric GAN for Unpaired Image-to-image Translation

Unpaired image-to-image translation problem aims to model the mapping fr...

Robustness via Uncertainty-aware Cycle Consistency

Unpaired image-to-image translation refers to learning inter-image-domai...

Digital twins of physical printing-imaging channel

In this paper, we address the problem of modeling a printing-imaging cha...

Please sign up or login with your details

Forgot password? Click here to reset