Uploaded March 2021 | Updated September 2026, 6 hours ago
Talk given to the UW Reality Lab, UW GRAIL, and the Graduate Vision Seminar by Hadar Elor
SPEAKER: Hadar Elor
TALK TITLE: Generation by Decomposition
ABSTRACT: Deep learning has revolutionized our ability to generate novel images and 3D shapes. Neural networks are typically trained to map a high-dimensional latent code to full realistic samples. In this talk, I will present two recent works focusing on generation of handwritten text and 3D shapes. In these works, we take a different approach and generate image and shape samples using a more granular part-based decomposition, demonstrating that the whole is not necessarily “greater than the sum of its parts”. I will also discuss how our generation by decomposition approach allows for a semantic manipulation of 3D shapes and improved handwritten text recognition performance.
BIO: Hadar Averbuch-Elor is a postdoctoral researcher at Cornell-Tech working with Prof. Noah Snavely. She completed her PhD in Electrical Engineering at Tel-Aviv University in Israel where she was advised by Prof. Daniel Cohen-Or. Her research focuses on modeling and manipulating visual concepts by combining pixels with more structured modalities, including natural language and 3D geometry. Hadar is supported by the Zuckerman STEM Leadership Program and the Schmidt Futures Program. http://www.cs.cornell.edu/~hadarelor/
Talk held and recorded over videochat on December 11th, 2020, and edited with the help of the UW CSE Production team (University of Washington, Computer Science and Engineering).
Talk given to the UW Reality Lab, UW GRAIL, and the Graduate Vision Seminar by Hadar Elor
SPEAKER: Hadar Elor
TALK TITLE: Generation by Decomposition
ABSTRACT: Deep learning has revolutionized our ability to generate novel images and 3D shapes. Neural networks are typically trained to map a high-dimensional latent code to full realistic samples. In this talk, I will present two recent works focusing on generation of handwritten text and 3D shapes. In these works, we take a different approach and generate image and shape samples using a more granular part-based decomposition, demonstrating that the whole is not necessarily “greater than the sum of its parts”. I will also discuss how our generation by decomposition approach allows for a semantic manipulation of 3D shapes and improved handwritten text recognition performance.
BIO: Hadar Averbuch-Elor is a postdoctoral researcher at Cornell-Tech working with Prof. Noah Snavely. She completed her PhD in Electrical Engineering at Tel-Aviv University in Israel where she was advised by Prof. Daniel Cohen-Or. Her research focuses on modeling and manipulating visual concepts by combining pixels with more structured modalities, including natural language and 3D geometry. Hadar is supported by the Zuckerman STEM Leadership Program and the Schmidt Futures Program. http://www.cs.cornell.edu/~hadarelor/
Talk held and recorded over videochat on December 11th, 2020, and edited with the help of the UW CSE Production team (University of Washington, Computer Science and Engineering).
