NYC Computer Vision Day drew hundreds of talented researchers to Brooklyn

The field of computer vision is a broad one, with practical applications across numerous sectors. Generally speaking, it involves developing algorithms and models that enable computers to “see” and interpret visual information. Medical practitioners can now harness the power of computer vision to make more accurate diagnoses, and manufacturers can leverage it to make their processes more efficient, resulting in potential cost savings for consumers. And if you’ve ever wondered how an autonomous car can possibly discern when a pedestrian crosses its path or how an easily downloadable app can recognize and translate a sign into a different language you’ve wondered about computer vision.

On February 3, more than 300 researchers from throughout the region converged on Brooklyn for the second annual NYC Computer Vision Day, an event organized by Assistant Professor David Fouhey, a faculty member of both NYU Tandon’s Department of Electrical and Computer Engineering and Courant’s Department of Computer Science.

The day brought together top graduate students and early career researchers from more than 75 university labs throughout the Northeast — including Princeton, the University of Maryland, Stony Brook, the University of Pennsylvania, Cornell, Brown, Columbia, the University of Massachusetts, Penn State, Johns Hopkins, Stevens Institute of Technology, Rutgers University, the University of Buffalo, the City University of New York, Yale, Fordham, University of Rochester, Temple, and Drexel — to hear keynote speakers and lightning talks and to view posters outlining work being done at the other institutions.

It’s an exciting time for computer vision. The researchers here today are shaping the future of the field and finding new applications with real-world benefits. I’m happy to be able to help create more community in the area and give early career researchers a chance to share their work.”
— Assistant Professor David Fouhey, organizer of NYC Computer Vision Day

Erin McGowan, a Ph.D. candidate at Tandon’s Visualization and Data Analytics (VIDA) Research Center, was there presenting the Center’s work on an Augmented Reality Guidance and User-Modeling System (ARGUS), an interactive visual analytics tool engineered to support the development of intelligent AR assistants that can run on devices like Microsoft HoloLens 2 or MagicLeap. She agreed with Fouhey that the opportunity to share ideas was invaluable. “As I viewed the posters, I saw researchers working on problems similar to the ones I’m working on,” she explains. “It was great to be able to confer and bounce ideas off of each other.”

There was also a chance to match faces with names previously seen only in the pages of journals. Arjun Krishna, a Ph.D. candidate at the University of Pennsylvania, had traveled to Brooklyn to present “The Value of Sensory Information to a Robot,” a paper that explores the question: what value does sensory information hold to a decision-making agent, such as a robot, at various moments in time during the execution of a given task? While he was excited to share those findings, he was even more excited to see the other presenters at the event. “The community-vision community is still relatively small, and in a sense, we know each other from reading the papers that emerge from other labs,” he says. “One of the best parts of being at NYC Computer Vision Day, for me, is to meet so many people whose work I’ve read and admired.”

Juan de Pablo, NYU’s Executive Vice President for Global Science and Technology and Executive Dean of Tandon, was on hand to welcome everyone to Brooklyn, and he emphasized how important the work of the computer-vision community is, to his own work as a materials scientist and chemical engineer and to the world as a whole. “Thanks to your work — and maybe even to new connections made today — we can look forward to a future of more effective medical care, more efficient manufacturing, safer transportation, and countless other improvements for humanity,” he asserted.

Fouhey stressed that NYC Computer Vision Day was only possible with the help of others. Among those owed thanks, he listed:

Advisory Committee Members: Carl Vondrick, Jia Deng, Olga Russakovsky

Program Committee Members: Mahi Shafiullah, Sarah Jabbour, Ruoshi Liu, Sunnie S.Y. Kim, Zhuang Liu

Keynote Speakers drawn from the newest computer-vision faculty in the region: Hadar Averbuch-Elor of Cornell, who presented “Editing 3D Objects with (and without) Multimodal Foundation Models,” and Yunzhu Li of Columbia, who presented “Foundation Models for Robotic Manipulation: Opportunities and Challenges.”

Corporate Sponsors: Adobe, Amazon, Google, Salesforce, and Snap Inc.