סמינר מחלקת מערכות - Dr. Hadar Averbuch-Elor - Learning About People from Text and Images
(The talk will be given in English)
Speaker: Dr. Hadar Averbuch-Elor
Department of EE-Systems, Tel Aviv University
011 hall, Electrical Engineering-Kitot Building
Monday, May 29th, 2023
15:00 - 16:00
Learning About People from Text and Images
Abstract
Interactions between humans are diverse and context-dependent, but previous works have treated them as categorical, disregarding the heavy tail of possible interactions. In this talk, I'll present a new paradigm of learning human-human interactions as free text from a single still image, allowing for more flexibility in modeling the unlimited space of situations and relationships between people. To overcome the absence of data labelled specifically for this task, we use knowledge distillation applied to synthetic caption data produced by a large language model without explicit supervision. I'll also introduce our person-centric image-caption dataset, automatically mined from Wikimedia Commons, which we also used for linking between people named in a caption and people pictured in an image.
Short Bio
Hadar Averbuch-Elor is an Assistant Professor at the School of Electrical Engineering in Tel Aviv University. Before that, Hadar was a postdoctoral researcher at Cornell-Tech. She completed her PhD in Electrical Engineering at Tel-Aviv University and is a recipient of several awards including the Zuckerman Postdoctoral Scholar Fellowship and the Alon Fellowship for the Integration of Outstanding Faculty. She also held research positions at Facebook and Amazon AI. Hadar's research interests lie in the intersection of computer graphics and computer vision.
השתתפות בסמינר תיתן קרדיט שמיעה = עפ"י רישום שם מלא + מספר ת.ז. בטופס הנוכחות שיועבר באולם במהלך הסמינר
