One-to-Many: Diverse and Bidirectional Image-to-Image Translation via Rectified Flow
Organizations: Kyushu University, Fukuoka, Japan · Current affiliation: CyberAgent, Inc.
Abstract
Many paired image-to-image translation tasks admit multiple valid target appearances, while training provides only one target per source. We propose One-to-Many Flow (O2MFlow), a latent-conditioned Rectified Flow (RF) model for one-to-many generation and backward translation (targetsource) within a single transport framework. Conditioning the RF dynamics on an appearance latent turns transport into a family of latent-indexed bijections: a fixed latent gives a mapping that is inverted by time-reversed integration, and sampled latents produce diverse outputs for the same input. We formalize reusable appearance variations as Shared Latent Transport and show that Kullback--Leibler (KL) regularization directly controls the source--latent distribution gap between training and generation. Across three benchmarks, O2MFlow offers dataset-dependent forward diversity--fidelity trade-offs and the lowest backward LPIPS-Dist among evaluated methods, with the lowest backward FID on two benchmarks.
Figures & tables
| Dataset /Method | LPIPS-Div | LPIPS-Dist | FID |
|---|---|---|---|
| edges2shoes | |||
| pix2pix | – | 0.1630 | 99.71 |
| CycleGAN | – | 0.2050 | 108.88 |
| BicycleGAN | 0.1914 | 0.2130 | 58.51 |
| DRIT++ | 0.1954 | 0.2280 | 50.18 |
| BBDM | 0.1089 | 0.1280 | 37.44 |
| Dataset /Method | LPIPS-Dist | FID |
|---|---|---|
| edges2shoes | ||
| pix2pix (baseline) | 0.0501 | 36.68 |
| CycleGAN | 0.2309 | 197.79 |
| SymmFlow | 0.0834 | 109.98 |
| BDBM | 0.0702 | 43.97 |
| O2MFlow | 0.0287 | 22.57 |
| Components | Forward | Backward | ||||
| Latent | KL | LPIPS-Div | LPIPS-Dist | FID | LPIPS-Dist | FID |
| edges2shoes | ||||||
| – | – | – | 0.1013 | 54.02 | 0.0277 | 23.09 |
| ✓ | – | 0.3269 | 0.3187 | 83.28 | 0.0281 | 23.36 |
| ✓ | ✓ | 0.1776 | 0.1924 | 50.86 | 0.0287 | 22.57 |
| edges2handbags | ||||||
Appendix figures & tables2 assets
Supplementary material from the paper’s appendix.
Appendix
| Steps | Fwd. LPIPS-Div | Fwd. FID | Bwd. FID |
|---|---|---|---|
| 5 | 0.1710 | 50.74 | 26.93 |
| 10 | 0.1721 | 50.74 | 23.21 |
| 25 | 0.1748 | 50.69 | 22.66 |
| 50 | 0.1776 | 50.86 | 22.57 |
Image-to-Image Translation
Image-to-image translation maps images between visual domains or modalities while preserving relevant content, structure, and semantics.
See all 122 papers