AI RESEARCH

Unified Vector Floorplan Generation via Markup Representation

arXiv CS.CV

ArXi:2604.04859v1 Announce Type: new Automatic residential floorplan generation has long been a central challenge bridging architecture and computer graphics, aiming to make spatial design efficient and accessible. While early methods based on constraint satisfaction or combinatorial optimization ensure feasibility, they lack diversity and flexibility. Recent generative models achieve promising results but struggle to generalize across heterogeneous conditional tasks, such as generation from site boundaries, room adjacency graphs, or partial layouts, due to their suboptimal representations.