University of Toronto

Current geometric design tools are indirect and require esoteric training. Ad hoc pipelines for preparing geometry for fabrication involve many phases and typically many different people or groups. Meanwhile, core geometry processing algorithms are not robust to data imperfections, which are now commonplace. Our approach is fundamentally different. We adapt fundamental mathematics to work with messy geometric data. An archetypical example is our past work on generalizing the classic formula for determining the inside from the outside of a curve to messy representations of 3D surface geometry commonly found throughout computer graphics and computer aided design. This work enables downstream processing such as physical simulation and fabrication.
We are pioneering large-scale testing of geometric algorithms. This empirical evidence of robust implementation complements correctness proofs. Robust core subroutines allows us to create complete geometric design systems connecting creation to fabrication in a single pipeline, similar to “what you see is what you get” word processing.
Lab Team
Eitan Grinspun
Columbia University
David I.W. Levin
University of Toronto
Maks Ovsjanikov
Ecole Polytechnique
Leonardo Sacht
Universidade Federal de Santa Catarina
Postdocs and Graduate Trainees
Seungbae Bang
Postdoctoral Fellow, University of Toronto
Thomas Davies
MSc Candidate, University of Toronto
Josh Holinaty
MSc Candidate, University of Toronto
Hsueh-Ti Derek Liu
PhD Candidate, University of Toronto
Sarah Kushner
PhD Candidate, University of Toronto
Silvia Sellán
PhD Candidate, University of Toronto
Risa Ulinski
PhD Candidate, University of Toronto