Latent Diffusion Models for Domain Adaptation

Finetuned semantic map-conditioned LDMs with ControlNet for unsupervised and unpaired Synthetic-to-Real image translation.

DATE August 2023

#ControlNet #Diffusion Models #Domain Adaptation

Latent Diffusion Models for Domain Adaptation

Overview

We explore the use of Latent Diffusion Models (LDMs) for domain adaptation, specifically from synthetic to real domains. By conditioning on semantic maps and utilizing ControlNet, we achieve realistic image translation without paired data, applicable for autonomous driving simulation and data augmentation.

CONTENTS

Latent Diffusion Models for Domain Adaptation

Overview