FAPE-IR：面向一体化图像复原的频率感知规划与执行框架
FAPE-IR: Frequency-Aware Planning and Execution Framework for All-in-One Image Restoration
将冻结的多模态大语言模型作为规划器生成频率感知的复原计划，并驱动基于LoRA的混合专家扩散执行器，为统一图像复原提供了可解释的语义规划与频率专家动态选择相结合的新范式。
原文依据 It uses a frozen Multimodal Large Language Model (MLLM) as a planner to analyze degraded images and generate concise, frequency-aware restoration plans.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论一体化图像复原（AIO-IR）旨在开发一个统一模型，能够处理复杂条件下的多种退化。然而，现有方法往往依赖任务特定的设计或潜变量路由策略，难以适应真实世界中存在各种退化的场景。我们提出了FAPE-IR，一个频率感知的规划与执行框架用于图像复原。它使用一个冻结的多模态大语言模型（MLLM）作为规划器，分析退化图像并生成简洁的频率感知复原计划。这些计划指导基于扩散模型的执行器中的LoRA混合专家（LoRA-MoE）模块，该模块动态选择高频或低频专家，并辅以输入图像的频率特征。为了进一步提高复原质量并减少伪影，我们引入了对抗训练和频率正则化损失。通过将语义规划与基于频率的复原相耦合，FAPE-IR为一体化图像复原提供了一个统一且可解释的解决方案。大量实验表明，FAPE-IR在七项复原任务上达到了最优性能，并在混合退化下展现出强大的零样本泛化能力。代码可在 https://github.com/Programmergg/FAPE-IR 获取。
英文原文摘要 1287 字符
All-in-One Image Restoration (AIO-IR) aims to develop a unified model that can handle multiple degradations under complex conditions. However, existing methods often rely on task-specific designs or latent routing strategies, making it hard to adapt to real-world scenarios with various degradations. We propose FAPE-IR, a Frequency-Aware Planning and Execution framework for image restoration. It uses a frozen Multimodal Large Language Model (MLLM) as a planner to analyze degraded images and generate concise, frequency-aware restoration plans. These plans guide a LoRA-based Mixture-of-Experts (LoRA-MoE) module within a diffusion-based executor, which dynamically selects high- or low-frequency experts, complemented by frequency features of the input image. To further improve restoration quality and reduce artifacts, we introduce adversarial training and a frequency regularization loss. By coupling semantic planning with frequency-based restoration, FAPE-IR offers a unified and interpretable solution for all-in-one image restoration. Extensive experiments show that FAPE-IR achieves state-of-the-art performance across seven restoration tasks and exhibits strong zero-shot generalization under mixed degradations. Code is available at https://github.com/Programmergg/FAPE-IR.