MetaPoint: Unlocking Precise Spatial Control in Agentic Visual Generation
2026-06-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors explain that current image-generating models have trouble placing objects exactly where you want on a picture. They introduce MetaPoint, which uses special tokens to represent exact spots on an image grid without changing the model's design. This lets the model control object positions precisely using just one or two tokens. MetaPoint also helps break down complex instructions into simpler parts, making it easier to create and edit images interactively.
generative visual modelsspatial controlpositional encodingtokenization2D coordinatesbounding boxcompositionalitygenerative agentsimage synthesisinteractive editing
Authors
Dewei Zhou, Xinyu Huang, Xun Wang, Ji Xie, Yabo Zhang, Liang Li, Kunchang Li, Zongxin Yang, Yi Yang
Abstract
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.