Match-and-Fuse：从无序图像集合进行一致性生成
Match-and-Fuse: Consistent Generation from Unstructured Image Sets
把无序图像集合建模为图、在成对生成时按稠密对应关系融合内部特征，这一免训练思路可直接迁移到多视图一致编辑、集合级个性化生成与视频帧间一致性约束，无需掩码或人工监督。
原文依据 fusing internal features across image pairs, guided by dense input correspondences, without requiring masks or manual supervision
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 Match-and-Fuse——一种零样本、免训练的方法，用于对无序图像集合进行一致的可控生成；这类集合共享某一共同视觉元素，但在视角、拍摄时间和周围内容上各不相同。现有方法通常作用于单张图像或密集采样的视频，与之不同，我们的框架执行集合到集合的生成：给定一个源集合与用户提示，它生成一个新的集合，并保持共享内容的跨图像一致性。我们的核心思想是将该任务建模为一张图，其中每个节点对应一张图像，每条边触发一对图像之间的联合生成。这一形式化将所有成对生成整合进统一框架，在施加局部一致性的同时确保整个集合的全局连贯性。具体而言，我们在稠密输入对应关系的引导下跨图像对融合内部特征，无需掩码或人工监督，并利用文本到图像模型中的一种涌现先验——当多个视图共享同一画布时，该先验会促使生成结果保持连贯。Match-and-Fuse 取得最优的一致性与视觉质量，并为基于图像集合的内容创作开辟了新的能力。
英文原文摘要 1235 字符
We present Match-and-Fuse - a zero-shot, training-free method for consistent controlled generation of unstructured image sets - collections that share a common visual element, yet differ in viewpoint, time of capture, and surrounding content. Unlike existing methods that operate on individual images or densely sampled videos, our framework performs set-to-set generation: given a source set and user prompts, it produces a new set that preserves cross-image consistency of shared content. Our key idea is to model the task as a graph, where each node corresponds to an image and each edge triggers a joint generation of image pairs. This formulation consolidates all pairwise generations into a unified framework, enforcing local consistency while ensuring global coherence across the entire set. This is achieved by fusing internal features across image pairs, guided by dense input correspondences, without requiring masks or manual supervision, and by leveraging an emergent prior in text-to-image models that encourages coherent generation when multiple views share a single canvas. Match-and-Fuse achieves state-of-the-art consistency and visual quality, and unlocks new capabilities for content creation from image collections.