arXiv 2610.135472026-10-12cs.CV

MOSAIC：面向遥感图像超分辨率的、基于图像引导原型条件的多尺度以对象为中心的槽适配

MOSAIC: Multi-Scale Object-Centric Slot Adaptation with Image-Guided Prototype Conditioning for Remote Sensing Image Super-Resolution

Zhengzhou University（郑州大学）；Aalto University（阿尔托大学）；Beijing University of Technology（北京工业大学）；Qilu University of Technology（齐鲁工业大学）

Liuqian Wang, Hanyu Xie, Rongzhen Zhao, Qi Ming, Chunjie Ma, Wentao Chao, Yufei Gao, Chengming Liu, Lei Shi

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究针对遥感图像超分辨率任务提出MOSAIC框架，通过以对象为中心的结构槽与共享原型内存适配冻结扩散Transformer，在UCMerced等数据集上实现PSNR与LPIPS指标提升，更好保留小对象等结构信息。

摘要：

预训练的扩散Transformer为图像复原提供了强大的生成先验，但将其适配到对结构敏感的任务仍具挑战性。现有方法主要依赖密集的低分辨率特征作为条件，这些特征常将显著对象与主导背景纠缠，且无法为跨尺度、跨样本的重复结构提供足够抽象性。我们提出MOSAIC，这是一个以对象为中心的表示学习框架，用于学习生成模型适配的紧凑结构条件。MOSAIC将退化观测组织为结构槽，以捕获互补的对象级和布局级证据，跨分辨率整合这些槽，并从共享原型内存中检索重复先验以补充不完整的图像特定信息。得到的表示通过轻量级残差调制注入到冻结扩散Transformer的选定层中，从而保留其预训练的生成先验。我们将MOSAIC实例化用于遥感图像超分辨率，其中小对象、重复结构和大规模变化构成了具有挑战性的测试平台。在UCMerced、AID和RSSCN7上的实验表明，在×2和×4设置下PSNR均有一致提升，且在×4设置下LPIPS结果更优。定性结果进一步证明其能更好地保留小对象、结构边界和重复空间模式。

英文摘要：

Pretrained diffusion Transformers provide powerful generative priors for image restoration, yet adapting them to structure-sensitive tasks remains challenging. Existing methods mainly rely on dense low-resolution features as conditions, which often entangle salient objects with dominant backgrounds and provide limited abstraction for recurring structures across scales and samples. We propose MOSAIC, an object-centric representation learning framework that learns compact structural conditions for generative model adaptation. MOSAIC organizes degraded observations into structural slots that capture complementary object-level and layout-level evidence, integrates them across resolutions, and retrieves recurring priors from a shared prototype memory to complement incomplete image-specific information. The resulting representations are injected into selected layers of a frozen diffusion Transformer through lightweight residual modulation, preserving its pretrained generative prior. We instantiate MOSAIC for remote sensing image super-resolution, where small objects, repetitive structures, and large-scale variation provide a challenging testbed. Experiments on UCMerced, AID, and RSSCN7 show consistent PSNR improvements across $\times 2$ and $\times 4$ settings and favorable LPIPS results at $\times 4$. Qualitative results further demonstrate improved preservation of small objects, structural boundaries, and repetitive spatial patterns.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。