DBMSolver: A Training-free Diffusion Bridge Sampler for High-Quality Image-to-Image Translation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913097553281024 |
|---|---|
| author | Venugopal, Sankarshana Mostafavi, Mohammad Choi, Jonghyun |
| author_facet | Venugopal, Sankarshana Mostafavi, Mohammad Choi, Jonghyun |
| contents | Diffusion-based image-to-image (I2I) translation excels in high-fidelity generation but suffers from slow sampling in state-of-the-art Diffusion Bridge Models (DBMs), often requiring dozens of function evaluations (NFEs). We introduce DBMSolver, a training-free sampler that exploits the semi-linear structure of DBM's underlying SDE and ODE via exponential integrators, yielding highly-efficient 1st- and 2nd-order solutions. This reduces NFEs by up to 5x while boosting quality (e.g., FID drops 53% on DIODE at 20 NFEs vs. 2nd-order baseline). Experiments on inpainting, stylization, and semantics-to-image tasks across resolutions up to 256x256 show DBMSolver sets new SOTA efficiency-quality tradeoffs, enabling real-world applicability. Our code is publicly available at https://github.com/snumprlab/dbmsolver. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_05889 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DBMSolver: A Training-free Diffusion Bridge Sampler for High-Quality Image-to-Image Translation Venugopal, Sankarshana Mostafavi, Mohammad Choi, Jonghyun Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Numerical Analysis Diffusion-based image-to-image (I2I) translation excels in high-fidelity generation but suffers from slow sampling in state-of-the-art Diffusion Bridge Models (DBMs), often requiring dozens of function evaluations (NFEs). We introduce DBMSolver, a training-free sampler that exploits the semi-linear structure of DBM's underlying SDE and ODE via exponential integrators, yielding highly-efficient 1st- and 2nd-order solutions. This reduces NFEs by up to 5x while boosting quality (e.g., FID drops 53% on DIODE at 20 NFEs vs. 2nd-order baseline). Experiments on inpainting, stylization, and semantics-to-image tasks across resolutions up to 256x256 show DBMSolver sets new SOTA efficiency-quality tradeoffs, enabling real-world applicability. Our code is publicly available at https://github.com/snumprlab/dbmsolver. |
| title | DBMSolver: A Training-free Diffusion Bridge Sampler for High-Quality Image-to-Image Translation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Numerical Analysis |
| url | https://arxiv.org/abs/2605.05889 |