UniCycleFlow：基于共享整流流的双向非配对图像翻译
UniCycleFlow: Bidirectional Unpaired Image Translation with a Shared Rectified Flow
浏览论文内容
中文总结 AI 辅助
UniCycleFlow是一种基于共享整流流的双向非配对图像翻译框架，通过将双向翻译纳入同一连续动态系统，在10项翻译任务中7项获最低FID，平均FID达55.1。
中文摘要 AI 辅助
双向非配对图像翻译需在无配对监督的情况下，学习两个方向的一致变换，同时保留源域特定结构。现有方法通常采用两个方向专用生成器，或训练独立的单向模型；即便通过循环一致性关联，这类模型仅约束往返端点重建，未要求两个方向遵循共同的局部变换规则。我们提出UniCycleFlow，一种整流流框架，将双向翻译表示为单个时间条件速度场的正向与反向积分，该表述将两个方向纳入同一连续动态系统，而非耦合原本独立的端点映射。核心挑战在于非配对数据无法提供构建整流流轨迹所需的有意义源-目标耦合，UniCycleFlow通过学习确定性源条件端点（其边际分布经对抗性匹配至相反域）解决此问题；所得路径经 stop-gradient 自流匹配（用于中间速度监督）、离散循环闭合（用于正-反向一致性）及表示路径-速度正则化（用于控制轨迹上的局部特征变化）进行正则化。在10个翻译方向上，UniCycleFlow在10项任务中的7项使用单次欧拉评估实现最低FID，且取得最佳平均FID为55.1。
英文摘要
Bidirectional unpaired image translation must preserve source-specific structure while learning coherent transformations in both directions without paired supervision. Existing methods typically employ two direction-specific generators or train separate one-way models. Even when linked by cycle consistency, such models constrain only the round-trip endpoint reconstruction, without requiring the two directions to obey a common local transformation rule. We propose UniCycleFlow, a rectified-flow framework that represents bidirectional translation as forward and reverse integration of a single time-conditioned velocity field. This formulation organizes both directions within the same continuous dynamics, rather than coupling otherwise separate endpoint mappings. A key challenge is that unpaired data provide no meaningful source--target coupling from which rectified-flow trajectories can be constructed. UniCycleFlow addresses this challenge by learning deterministic source-conditioned endpoints whose marginal distributions are adversarially matched to the opposite domains. The resulting paths are regularized by stop-gradient self-flow matching for intermediate velocity supervision, discrete cycle closure for forward--reverse consistency, and representation path-velocity regularization for controlling localized feature changes along the trajectory. Across ten translation directions, UniCycleFlow achieves the lowest FID on 7 of 10 tasks using a single Euler evaluation and obtains the best average FID of 55.1.
