解耦残差去噪扩散模型：面向统一且数据高效的图像到图像翻译
Decoupled Residual Denoising Diffusion Models for Unified and Data Efficient Image-to-Image Translation
把扩散过程拆成「随机噪声扩散做域对齐+流形提升」与「确定性残差扩散学语义映射」两段，且噪声阶段只用大量无配对目标域图像训练，为少配对数据的统一 I2I 翻译提供了可复用的训练范式。
原文依据 DRDD decouples the diffusion process into two sequential and independent diffusion stages
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出解耦残差去噪扩散模型（DRDD），用于统一且数据高效的图像到图像（I2I）翻译。尽管扩散模型在质量和多样性方面推动了 I2I 翻译的发展，我们揭示了扩散模型中一个此前未被充分探索的性质。关键在于，除了流形提升（即把数据移离低维流形）这一传统作用之外，注入高斯噪声还能通过隐式对齐跨域的特征分布来促进域协调，这一性质对统一 I2I 翻译尤为有利。然而，现有扩散模型会过早地侵蚀这种协调效应，因为噪声与残差在单一的耦合扩散过程中被同时去除。为解决这一问题，DRDD 将扩散过程解耦为两个顺序且独立的扩散阶段：（1）随机噪声扩散，用于域协调与流形提升；（2）确定性残差扩散，在固定噪声域内完整地学习核心语义映射。这种解耦在整个变换过程中保留了协调与流形提升效应，大幅简化了跨多样任务与域的统一映射的学习。值得注意的是，噪声扩散阶段仅使用大量无配对的目标域图像进行训练，显著提升了数据效率。全面的理论与实证分析表明，DRDD 与主流扩散模型广泛兼容，即使在配对数据有限的情况下也能稳定地实现稳健、统一的 I2I 翻译。我们的代码已发布于 https://github.com/HKU-HealthAI/DRDD。
英文原文摘要 1670 字符
We propose Decoupled Residual Denoising Diffusion models (DRDD) for unified and data-efficient image-to-image (I2I) translation. While diffusion models have advanced I2I translation in terms of quality and diversity, we uncover a previously under-explored property in diffusion models. Crucially, beyond its conventional role of manifold lifting (i.e., moving data off low-dimensional manifolds), injecting Gaussian noise facilitates domain harmonization by implicitly aligning feature distributions across domains, a property particularly advantageous for unified I2I translation. However, existing diffusion models prematurely erode this harmonization effect, as noise and residuals are simultaneously removed in a single coupled diffusion process. To address this, DRDD decouples the diffusion process into two sequential and independent diffusion stages: (1) a stochastic noise diffusion for domain harmonization and manifold lifting, and (2) a deterministic residual diffusion that learns the core semantic mapping entirely within the fixed-noise domain. This decoupling preserves harmonization and manifold lifting effects throughout the transformation, substantially simplifying the learning of unified mappings across diverse tasks and domains. Notably, the noise diffusion stage is trained exclusively on abundant, unpaired target-domain images, greatly improving data efficiency. Comprehensive theoretical and empirical analysis demonstrates that DRDD is broadly compatible with mainstream diffusion models and consistently delivers robust, unified I2I translation, even under limited paired data. Our code is available at https://github.com/HKU-HealthAI/DRDD.