Jun-Yan is a Ph.D. student at EECS Department of UC Berkeley. Before moving to Berkeley, he was a Computer Science Ph.D. student at CMU. He received his bachelor degree from Tsinghua University. He is now working on computer graphics, computer vision and data mining with Professor Alexei A. Efros. In particular, his research focuses on summarizing, mining and exploring large-scale visual data collections, with the goal of building a digital bridge between Humans and Big Visual Data.

Research Summary

We are living in an age of Big Visual Data. An estimated 3.5 trillion photographs have been taken since the invention of photography, of which 10% within the past 12 months. Facebook alone reports 9 billion photo uploads per month. In fact, there is so much visual data out there that much of it might never be seen by a human being! But unlike other types of “Big Data” (e.g. text), much of the visual content cannot be easily indexed, searched or hyperlinked, making it Internet’s “digital dark matter”. How can we help humans better understand this vast visual space, to see what’s out there? Jun-Yan’s research is directed towards connecting humans to huge amounts of unorganized images and videos. He has proposed methods for automatically mining visual patterns including both the visual commonality in a photo collection, and the visual difference between individual items. Based on the discovered visual patterns, he develops human-machine visual interface to communicate user’s mental picture directly to the visual data using a palette of visual communication input tools. He is now working on AverageExplorer, an interactive framework that allows a user to rapidly explore and visualize a large image collection using the medium of average images. For more information, please visit his website: http://www.eecs.berkeley.edu/~junyanz/