Single-Step Bidirectional Unpaired Image Translation Using Implicit Bridge Consistency Distillation
This addresses a practical bottleneck for real-world applications of diffusion-based translation methods, though it is incremental over existing consistency distillation techniques.
The paper tackles the challenge of slow iterative sampling in unpaired image-to-image translation by proposing Implicit Bridge Consistency Distillation (IBCD), which enables single-step bidirectional translation without adversarial loss, achieving state-of-the-art performance on benchmark datasets.
Unpaired image-to-image translation has seen significant progress since the introduction of CycleGAN. However, methods based on diffusion models or Schrödinger bridges have yet to be widely adopted in real-world applications due to their iterative sampling nature. To address this challenge, we propose a novel framework, Implicit Bridge Consistency Distillation (IBCD), which enables single-step bidirectional unpaired translation without using adversarial loss. IBCD extends consistency distillation by using a diffusion implicit bridge model that connects PF-ODE trajectories between distributions. Additionally, we introduce two key improvements: 1) distribution matching for consistency distillation and 2) adaptive weighting method based on distillation difficulty. Experimental results demonstrate that IBCD achieves state-of-the-art performance on benchmark datasets in a single generation step. Project page available at https://hyn2028.github.io/project_page/IBCD/index.html