I2E: 从图像像素到可操作的交互环境：用于文本引导的图像编辑
I2E: From Image Pixels to Actionable Interactive Environments for Text-Guided Image Editing
AI总结：
I2E提出一种'分解-然后-行动'范式，通过分解器将无结构图像转换为可操作的对象层，并利用物理感知的视觉-语言-行动代理进行复杂指令解析，同时构建I2E-Bench基准测试集，实验表明其在处理复杂组合指令和保持物理合理性方面优于现有方法。
摘要：
现有的文本引导图像编辑方法主要依赖端到端的像素级修补范式。尽管在简单场景中取得成功，但该范式在需要精确局部控制和复杂多对象空间推理的组合编辑任务中仍面临显著挑战。该范式受到三个限制：1) 计划与执行的隐含耦合，2) 缺乏对象级控制粒度，3) 依赖无结构的像素中心建模。为了解决这些限制，我们提出了I2E，一种新的'分解-然后-行动'范式，重新审视图像编辑作为结构化环境中的可操作交互过程。I2E利用分解器将无结构图像转换为离散、可操作的对象层，然后引入一个物理感知的视觉-语言-行动代理，通过链式推理将复杂指令解析为一系列原子动作。此外，我们还构建了I2E-Bench，一个用于多实例空间推理和高精度编辑的基准测试集。在I2E-Bench和多个公开基准上的实验结果表明，I2E在处理复杂组合指令、保持物理合理性以及确保多轮编辑稳定性方面显著优于现有方法。
英文摘要：
Existing text-guided image editing methods primarily rely on end-to-end pixel-level inpainting paradigm. Despite its success in simple scenarios, this paradigm still significantly struggles with compositional editing tasks that require precise local control and complex multi-object spatial reasoning. This paradigm is severely limited by 1) the implicit coupling of planning and execution, 2) the lack of object-level control granularity, and 3) the reliance on unstructured, pixel-centric modeling. To address these limitations, we propose I2E, a novel "Decompose-then-Action" paradigm that revisits image editing as an actionable interaction process within a structured environment. I2E utilizes a Decomposer to transform unstructured images into discrete, manipulable object layers and then introduces a physics-aware Vision-Language-Action Agent to parse complex instructions into a series of atomic actions via Chain-of-Thought reasoning. Further, we also construct I2E-Bench, a benchmark designed for multi-instance spatial reasoning and high-precision editing. Experimental results on I2E-Bench and multiple public benchmarks demonstrate that I2E significantly outperforms state-of-the-art methods in handling complex compositional instructions, maintaining physical plausibility, and ensuring multi-turn editing stability.
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。
