UniReal：通过学习真实世界动态实现通用图像生成与编辑
UniReal: Universal Image Generation and Editing via Learning Real-world Dynamics
AI总结：
本研究提出统一框架UniReal，将图像生成与编辑任务视为非连续视频生成，以大规模视频为监督学习真实世界动态，可支持多种图像任务并具备新应用的涌现能力。
摘要：
我们推出UniReal，这是一个旨在处理各类图像生成与编辑任务的统一框架。现有解决方案通常因任务不同而有差异，但都遵循基本原则：在捕捉视觉变化的同时保持输入与输出之间的一致性。受近期能有效平衡帧间一致性与变化性的视频生成模型启发，我们提出一种统一方法，将图像级任务视为非连续视频生成。具体而言，我们将数量可变的输入和输出图像视为帧，从而无缝支持图像生成、编辑、定制、合成等任务。尽管是为图像级任务设计的，我们仍将视频作为通用监督的可扩展来源。UniReal从大规模视频中学习世界动态，在处理阴影、反射、姿态变化和物体交互方面展现出先进能力，同时还具备应对新型应用的涌现能力。
英文摘要：
We introduce UniReal, a unified framework designed to address various image generation and editing tasks. Existing solutions often vary by tasks, yet share fundamental principles: preserving consistency between inputs and outputs while capturing visual variations. Inspired by recent video generation models that effectively balance consistency and variation across frames, we propose a unifying approach that treats image-level tasks as discontinuous video generation. Specifically, we treat varying numbers of input and output images as frames, enabling seamless support for tasks such as image generation, editing, customization, composition, etc. Although designed for image-level tasks, we leverage videos as a scalable source for universal supervision. UniReal learns world dynamics from large-scale videos, demonstrating advanced capability in handling shadows, reflections, pose variation, and object interaction, while also exhibiting emergent capability for novel applications.
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。
