What's So Hard About Big Visual Data?
With an estimated 3.5 trillion photographs in the world today, Cisco estimates that in the next few years, visual data (photos and video) will acount for over 85% of total internet traffic. Surprisingly, however, we currently lack effective computational methods for making sense of this mass of visual data, referred to as the Internet’s “digital dark matter.” In this talk, Alexei Efros of Carnegie Mellon University will discuss some of the unique challenges that make Big Visual Data difficult to interpret compared to other types of content. Efros also presents some recent work done by Carnegie Mellon University that aims to address this challenge in the context of visual matching, image retrieval, and visual data mining.
Alexei Efros, Carnegie Mellon University