Studying people to improve computer vision
David Crandall, Indiana University
3:30 p.m.–4:45 p.m., September 22, 2022 | 140 DeBartolo Hall
While early work in computer vision was inspired by studies of human perception, most recent work has focused on techniques that work well in practice but probably have little biological basis. But low-cost, lightweight wearable cameras and gaze trackers can now record people’s actual fields of view as they go about their everyday lives. Such first-person, “egocentric” video contains rich information about how people see and interact with the world around them, potentially helping us better understand human perception and behavior while also yielding insights that could improve computer vision. For example, studying how young children interact with unfamiliar toys could help computer vision researchers design better techniques for learning computational object models.
In this talk, I’ll describe several recent interdisciplinary projects in which we have used computer vision to study and model people’s behavior from a first-person perspective and then used these insights to try to improve computer vision. I’ll also talk about a project to collect a large-scale dataset of egocentric video data to push forward work in this area.
Finally, I’ll talk about recent work in which we studied the people in the computer vision community itself, trying to understand how researchers and practitioners feel about the trajectory of the field and how to improve it.
David Crandall is the Luddy Professor of Computer Science at Indiana University, where he is also the Inaugural Director of the Luddy Center for Artificial Intelligence, and Director of Graduate Studies for Computer Science.
He obtained the Ph.D. in Computer Science from Cornell University and B.S. and M.S. degrees in Computer Science and Engineering from the Pennsylvania State University. He was also a Senior Research Scientist at Eastman Kodak Company from 2001-2003. He is an Associate Editor of the IEEE Transactions on Pattern Analysis and Machine Intelligence and the IEEE Transactions on Multimedia.
In addition to numerous best paper awards, he has received an NSF CAREER award (2013), two Google Faculty Research Awards (2014 and 2020), an IU Trustees Teaching Award (2017), a Grant Thornton Fellowship (2019), and a Luddy Named Professorship (2021).