Volumetric Design

Volumetric design, or massing design, focuses on the initial 3D shape creation in architectural and engineering design, aiming to automate and improve the efficiency of this crucial early stage. Current research emphasizes learning sequential design processes using deep learning models, such as transformers and autoencoders, often applied to large datasets of existing designs to learn design preferences and generate new designs. These methods leverage techniques like clustering to handle high-resolution data and improve the quality and efficiency of design generation and evaluation, impacting both the speed and quality of design workflows.