SketchOpt: Sketch-based Parametric Model Retrieval for Generative Design

Developing fully parametric building models for performance-based generative design tasks often requires proficiency in many advanced 3D modeling and visual programming software, limiting its use for many building designers. Moreover, iterations of such models can be time-consuming tasks and sometimes limiting depending on the the design stage, as major changes in the layout design may result in remodeling the entire parametric definition. To address these challenges, we introduce a novel automated generative design system, which takes a basic floor plan sketch as an input and provides a parametric model prepared for multi-objective building optimization as output. In addition, the user-designer can assign various design variables for its desired building elements by using simple annotations in the drawing. We take advantage of a asymmetric convolutional module combined with a parametrizer to allow real-time parametric sketch-retrieval for a performance-based generative workflow. The system would recognize the corresponding element and define variable constraints to prepare for a multi-objective optimization problem. We illustrate the the use case of our proposed system by running a real-time structural optimization form-finding study. Our findings indicate the system can be utilized as a promising generative design tool for novice users.