Shiry Ginosar

University of California, Berkeley

PhD Candidate

I am a PhD candidate in Computer Vision at the CS Department of the University of California Berkeley working with Professor Alyosha Efros.

Previously I was a Human Computer Interaction researcher. I spent some time as a Visiting Scholar at the CS Department of Carnegie Mellon University, working with Professor Luis von Ahn in the field of Human Computation. More recently, I was part of Professor Bjoern Hartmann’s lab at Berkeley. In between I spent four years at Endeca as a Senior Software Engineer.

My work has been covered by The New Yorker, The Wall Street Journal, and the Washington Post among many others. It was exhibited at the Israeli Design Museum and is part of the permanent collection of the Deutsches Museum.

I am a recipient of the U.S. National Science Foundation Graduate Research Fellowship, the California Legislature Grant for graduate studies and the Samuel Silver Memorial Scholarship Award for combining intellectual achievement in science and engineering with serious humanistic and cultural interests.

Research Abstract:

Computers still represent all humans as a generic person class. But actually, real humans are full of rich subtle details. My goal is to build systems that go beyond categorical word-label representations and can mine for correlations, discovering the rich subtleties of real life from large quantities of raw data such as images, video and audio. As steps towards this vision, my research has focused on discovering indescribable things: how people and behavior change over time, the small details of life-like appearances and the nuances of person specific motion. Since I seek to find complex and high dimensional correlations, I have taken a learning approach from large image and video datasets without detailed human annotations. This is in contrast to learning from human class labels and traditional physics-based modeling approaches that fail to capture the richness of real humans.