WallPlan: synthesizing floorplans by learning to generate wall graphs

Floorplan generation has drawn widespread interest in the community. Re- cent learning-based methods for generating realistic floorplans have made significant progress while a complex heuristic post-processing is still neces- sary to obtain desired results. In this paper, we propose a novel wall-oriented method, called WallPlan , for automatically and efficiently generating plausi- blefloorplansfromvariousdesignconstraints.Wepioneertherepresentation ofthefloorplanasawallgraphwithroomlabelsandconsiderthefloorplangenerationasagraphgeneration.Giventheboundaryasinput,wefirst initializetheboundarywithwindowspredictedbyWinNet.ThenagraphgenerationnetworkGraphNetandsemanticspredictionnetworkLabelNet arecoupledtogeneratethewallgraphprogressivelybyimitatinggraphtra-versal. WallPlan can be applied for practical architectural designs, especially the wall-based constraints. We conduct ablation experiments, qualitative evaluations, quantitative comparisons, and perceptual studies to evaluate our method’s feasibility, efficacy, and versatility. Intensive experiments demon- strate our method requires no post-processing, producing higher quality floorplans than state-of-the-art techniques.