Constrained Layout Generation with Factor Graphs

TL;DR

This paper introduces a factor graph neural network approach for object-centric layout generation under spatial constraints, improving accuracy and suitability for iterative human-in-the-loop design processes.

Contribution

It presents a novel factor graph-based method with message passing for layout generation, capturing complex object interactions more effectively than previous single-node models.

Findings

Significant improvement in IOU scores over existing methods.

Effective modeling of higher-order spatial constraints.

Suitable for iterative human-in-the-loop design workflows.

Abstract

This paper addresses the challenge of object-centric layout generation under spatial constraints, seen in multiple domains including floorplan design process. The design process typically involves specifying a set of spatial constraints that include object attributes like size and inter-object relations such as relative positioning. Existing works, which typically represent objects as single nodes, lack the granularity to accurately model complex interactions between objects. For instance, often only certain parts of an object, like a room's right wall, interact with adjacent objects. To address this gap, we introduce a factor graph based approach with four latent variable nodes for each room, and a factor node for each constraint. The factor nodes represent dependencies among the variables to which they are connected, effectively capturing constraints that are potentially of a higher order. We then develop message-passing on the bipartite graph, forming a factor graph neural network that is trained to produce a floorplan that aligns with the desired requirements. Our approach is simple and generates layouts faithful to the user requirements, demonstrated by a large improvement in IOU scores over existing methods. Additionally, our approach, being inferential and accurate, is well-suited to the practical human-in-the-loop design process where specifications evolve iteratively, offering a practical and powerful tool for AI-guided design.