Fashion meets Technology: A Match made in the Big Data World
Nov 18, 2013
from 04:00 PM to 05:00 PM
|Where||Engr. IV Bldg., Maxwell Room 57-124|
|Contact Name||Prof. Vwani Roychowdhury|
|Add event to calendar||
Dr. Sudhir Singh
In this talk, I will start with reviewing the various phases of data evolution on the Internet, in particular the rise of the visual data. Then I will outline a new framework for learning from the Big Visual Data to tackle one of the most challenging problems of Al - to enable machines see. In particular, I will discuss the technology behind StileEye - a visual fashion discovery platform. The $300B fashion industry presents an interesting challenge - how to make computers see fashion on the Web exactly as consumers see it online and in everyday life? To do this, one has to isolate and identify clothing and accessories within visually complex images like personal photos and magazine spreads. 40 years of research in the field of computer vision and machine learning falls flat when put to a test with the learning from the Internet scale data. There is an inherent tradeoff between speed, accuracy and context, and at Internet scale it forces us to revisit even the fundamental aspects of computer vision such as color and shape beyond the state-of-art. I will present a new framework that rebuilds these fundamentals as well as teaches how to learn from Internet scale visual data on top of these to enable a machine to see and understand.
Sudhir Kumar Singh is co-founder and VP of Technology at StileEye, wherein he leads the conception, design and development of its revolutionary visual fashion discovery platform. He has contributed significantly to various fundamental emerging aspects of information and computation such as Quantum Computing, Algorithmic Game Theory, Network Information Theory, and most recently Large Scale machine learning and Computer vision. Dr. Singh holds a Ph.D. in Electrical Engineering from UCLA.