Casa Creator

Abstract

CasaCreator is another project that focuses on automating the generation of residential floor plans using advanced deep learning techniques. Using a large data set of more than 80,000 real-world residential floor plans, our method uses a two-dimensional neural network. In the first stage, the system predicts building locations, while in the second stage, it refines the system with pixel-level accuracy of walls. This new approach ensures a true and realistic floor plan. Our work finds applications in a variety of industries, including construction, real estate, interior design and in addition to this concept, it can also be applied to the gaming industry where the virtual environment of the gaming environment may require it floor plans for virtual buildings. Architects and designers can use CasaCreator to quickly prototype residential design, enabling more efficient use of space and creative exploration. The designs are consistent with defined product boundaries and constraints, and demonstrate the project’s ability to be successfully applied in realworld contexts CasaCreator stands as a testament to innovative technology and real-world applications integrated, and transforming the way residential floor plans are conceived and produced.