Sketch and Extrude

Sketch-and-extrude is a fundamental process in computer-aided design (CAD) where 2D sketches are used to define the cross-section of a 3D object, which is then extruded along a path to create the final shape. Current research focuses on automating this process using machine learning, particularly employing neural networks like Transformers and diffusion models to reconstruct 3D models from various inputs, including multi-view images, orthographic drawings, and even point clouds. These advancements aim to improve the efficiency and accuracy of 3D modeling, impacting fields such as robotics, CAD software development, and digital art creation by enabling more intuitive and automated design workflows.