Generative AI for 3D (3D AI) - Prof Rana Hanocka
About the Speaker
Rana Hanocka is the Founder and CEO of Thrixel and a Professor of Computer Science at the University of Chicago. She is a global leader in the field of 3D AI, with over a decade of research at the intersection of geometry processing and machine learning. Her works, including MeshCNN, Text2Mesh, and LL3M, have introduced foundational methods for learning directly on 3D data, with broad impact across computer graphics, vision, and generative AI.
Much of the current success of deep learning has been driven by massive amounts of curated data, whether annotated or unannotated. Compared to image datasets, developing large-scale 3D datasets is either prohibitively expensive or impractical. In this talk, Prof. Hanocka will present several works that harness the power of data-driven deep learning for tasks in shape editing and processing, without any 3D datasets. She will discuss works that learn to synthesize and analyze 3D geometry using large image datasets.
Selected highlights of her work:
MeshCNN: She pioneered the development of MeshCNN, a specialized convolutional neural network designed specifically for 3D triangular meshes.
Text-to-3D Innovation: Her recent NSF-backed project, 3DStylus, allows users to edit and create complex 3D shapes using only natural language (e.g., transforming a 3D model with a prompt like "colorful crochet candle").
Award-Winning Research: A recipient of the 2025 Best Paper Honorable Mention at 3DV and a former "Rising Star in EECS," Professor Hanocka is at the forefront of moving AI beyond 2D images into the unstructured world of 3D geometry.