Abstract

Today's design environments take advantage of powerful computational technology more than ever before. However, as design software becomes increasingly more sophisticated, it also becomes more difficult to master, requiring an increased demand for human specialization and expertise. In the Visual Design and Engineering Lab at Carnegie Mellon, we are developing human-centric 3D modeling techniques to assist shape creation in the early stages of the design process. Our fundamental strategy is to enable a level of design abstraction to appropriately support visual thinking and computational ideation activities.

In this talk, I will highlight our research in this area with an emphasis on natural user interfaces for shape creation and manipulation. In the first half the talk, I will present a new modeling software for rapidly designing and detailing 3D shapes. In the second half, I will describe a data-driven shape analysis technique that can geometrically abstract existing shapes, and use the results to decipher tacit design rules underlying man-made shapes. Based on this technique, I will demonstrate our recent work in co-constrained shape manipulation and brand-identity recognition. Finally, I will describe our ongoing work in semantically driven CAD technology, where shape editing is controlled by perceptual design attributes rather than conventional geometric handles.

Bio

Levent Burak Kara is an associate professor of Mechanical Engineering at Carnegie Mellon University. He is the founder of Visual Design and Engineering Laboratory. His research interests include computer-aided design, computer graphics, natural user interfaces and machine learning, with applications in industrial product design, automotive design, engineering education and bio-medical engineering. He is the recipient of National Science Foundation Career award and American Society of Mechanical Engineers Design Automation Society Young Investigator Award. At CMU, he teaches the Mechanical Engineering Senior Capstone Design course and the graduate level AI and Machine Learning for Engineering Design course. Kara has a BS in Mechanical Engineering from the Middle East Technical University, and an MS and PhD in Mechanical Engineering from Carnegie Mellon University.