Dual Diffusion Implicit Bridges for Image-to-Image Translation

by   Xuan Su, et al.
Stanford University

Common image-to-image translation methods rely on joint training over data from both source and target domains. This excludes cases where domain data is private (e.g., in a federated setting), and often means that a new model has to be trained for a new pair of domains. We present Dual Diffusion Implicit Bridges (DDIBs), an image translation method based on diffusion models, that circumvents training on domain pairs. DDIBs allow translations between arbitrary pairs of source-target domains, given independently trained diffusion models on the respective domains. Image translation with DDIBs is a two-step process: DDIBs first obtain latent encodings for source images with the source diffusion model, and next decode such encodings using the target model to construct target images. Moreover, DDIBs enable cycle-consistency by default and is theoretically connected to optimal transport. Experimentally, we apply DDIBs on a variety of synthetic and high-resolution image datasets, demonstrating their utility in example-guided color transfer, image-to-image translation as well as their connections to optimal transport methods.


page 2

page 8

page 9

page 10

page 12

page 13


Extremal Domain Translation with Neural Optimal Transport

We propose the extremal transport (ET) which is a mathematical formaliza...

Unpaired Image Translation via Vector Symbolic Architectures

Image-to-image translation has played an important role in enabling synt...

Leveraging Local Domains for Image-to-Image Translation

Image-to-image (i2i) networks struggle to capture local changes because ...

EGSDE: Unpaired Image-to-Image Translation via Energy-Guided Stochastic Differential Equations

Score-based diffusion generative models (SDGMs) have achieved the SOTA F...

Attribute-Driven Spontaneous Motion in Unpaired Image Translation

Current image translation methods, albeit effective to produce high-qual...

CoMoGAN: continuous model-guided image-to-image translation

CoMoGAN is a continuous GAN relying on the unsupervised reorganization o...

Image-to-Image Translation with Multi-Path Consistency Regularization

Image translation across different domains has attracted much attention ...

Please sign up or login with your details

Forgot password? Click here to reset