用一个特殊标记实现像素级空间控制，让视觉生成更精准
MetaPoint: Unlocking Precise Spatial Control in Agentic Visual Generation
- 用单个令牌表示二维坐标，直接利用模型原有位置编码
- 仅用1个令牌即可精确控制物体位置，2个令牌控制边界框
- 适合需要精细布局的生成代理和交互式编辑系统
生成式视觉模型在精确空间控制上存在根本性困难，根源在于模型能理解文本描述的空间关系，却无法将数值坐标映射到二维图像画布。本文提出MetaPoint方法，通过将连续二维坐标表示为单一特殊标记，直接利用模型固有的位置编码机制进行解析，将该标记视为画布上的虚拟点。该轻量级方法无需新增架构组件，也无需定制注意力掩码，仅用一个令牌即可实现像素级物体定位，两个令牌可控制其边界框。这些元点标记具有组合性，可作为空间基本单元，使规划代理将高层用户指令分解为结构化操作序列。MetaPoint提供了一种简单、精确且可扩展的空间控制基础，推动更强大的组合式生成代理发展，并支持直观的交互式编辑系统。
原文摘要 · Abstract (English)
Generative visual models fundamentally struggle with precise spatial control. This arises from a core disconnect: models can process textual descriptions of space but cannot directly map numerical coordinates onto the 2D image canvas. We introduce MetaPoint, a method that bridges this gap by representing a continuous 2D coordinate as a single, special token. Crucially, MetaPoint requires no new architectural components; it directly leverages the model's inherent positional encoding schemes to interpret these coordinates, treating our token as a virtual point on the canvas. This lightweight approach enables pixel-level control of an object's position with one token or its bounding box with two, all without requiring architectural changes or bespoke attention masking. The MetaPoint tokens are designed to be compositional, serving as spatial primitives. This allows a planner agent to decompose a high-level user request into a structured sequence of primitives for the generator. By providing a simple, precise, and scalable building block for spatial control, MetaPoint unlocks more powerful compositional generative agents and enables intuitive, interactive editing systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。