以你的方式融合：通过直接偏好优化使图像融合对齐异构需求
Fusion in Your Way: Aligning Image Fusion with Heterogeneous Demands via Direct Preference Optimization
把 DPO 引入潜扩散模型的采样过程，用偏好信号直接控制生成结果的属性，为视觉中「同一输入需输出多种风格/质量取向结果」的可控生成提供了一套可复用的偏好对齐训练范式。
原文依据 we propose DPOFusion, a direct preference optimization (DPO) framework integrating the property-aligned latent diffusion model (PALDM)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论作为多模态处理中的关键技术，红外与可见光图像融合（IVIF）在整合互补光谱信息以增强视觉表现并服务下游视觉任务方面发挥着重要作用。尽管已取得显著进展，现有方法仍难以灵活满足异构需求。实现与人类视觉和机器视觉多种偏好相一致的自适应融合，仍是一个开放且具有挑战性的问题。为应对这一挑战，我们提出 DPOFusion，一个直接偏好优化（DPO）框架，集成属性对齐潜扩散模型（PALDM）与偏好可控潜扩散模型（PCLDM），从而实现面向人类视觉与机器视觉的、任务引导且偏好自适应的 IVIF。PALDM 利用潜变量融合先验与联合条件损失，生成具有多种属性的多样化候选融合结果。随后 PCLDM 通过实例级直接偏好优化（IDPO）进行微调，从而能够以异构偏好信号直接控制最终融合结果。实验结果表明，我们的框架不仅在人类、视觉语言模型与任务驱动网络之间实现了精确的偏好对齐，还为自适应融合质量与面向任务的迁移能力树立了新的基准。我们的代码将发布于 https://github.com/suweijian1996/DPOFusion。
英文原文摘要 1402 字符
As a key technique in multi-modal processing, infrared and visible image fusion (IVIF) plays a crucial role in integrating complementary spectral information for visual enhancement and downstream vision tasks. Despite remarkable progress, existing methods struggle to flexibly accommodate heterogeneous demands. Achieving adaptive fusion that aligns with various preferences from both human and machine vision remains an open and challenging problem. To address this challenge, we propose DPOFusion, a direct preference optimization (DPO) framework integrating the property-aligned latent diffusion model (PALDM) and the preference-controllable latent diffusion model (PCLDM), enabling task-guided, preference-adaptive IVIF for both human and machine vision. The PALDM leverages a latent fusion prior and a joint conditional loss to generate diverse candidate fusion results with various properties. PCLDM is subsequently fine-tuned via instance direct preference optimization (IDPO), enabling direct control of the final fusion results with heterogeneous preference signals.Experimental results demonstrate that our framework not only attains precise preference alignment among humans, vision-language models, and task-driven networks, but also sets a new benchmark for adaptive fusion quality and task-oriented transferability. Our code will be available at https://github.com/suweijian1996/DPOFusion