BBDM: Image-to-image Translation with Brownian Bridge Diffusion Models

TL;DR

This paper introduces BBDM, a novel image-to-image translation method using Brownian Bridge diffusion models, which models translation as a stochastic process, outperforming traditional conditional diffusion approaches.

Contribution

The paper presents the first application of Brownian Bridge diffusion processes for image-to-image translation, enabling direct domain translation through bidirectional diffusion.

Findings

Achieves competitive performance on various benchmarks.

Models translation as a stochastic Brownian bridge process.

Outperforms existing conditional diffusion models.

Abstract

Image-to-image translation is an important and challenging problem in computer vision and image processing. Diffusion models (DM) have shown great potentials for high-quality image synthesis, and have gained competitive performance on the task of image-to-image translation. However, most of the existing diffusion models treat image-to-image translation as conditional generation processes, and suffer heavily from the gap between distinct domains. In this paper, a novel image-to-image translation method based on the Brownian Bridge Diffusion Model (BBDM) is proposed, which models image-to-image translation as a stochastic Brownian bridge process, and learns the translation between two domains directly through the bidirectional diffusion process rather than a conditional generation process. To the best of our knowledge, it is the first work that proposes Brownian Bridge diffusion process for image-to-image translation. Experimental results on various benchmarks demonstrate that the proposed BBDM model achieves competitive performance through both visual inspection and measurable metrics.