Liu part of Google team that won computer vision challenge

Wei Liu (photo by Brett Piper).
Wei Liu (photo by Brett Piper).

Computer science doctoral student Wei Liu was part of a Google team that won two of six categories in the worldwide ImageNet Large Scale Visual Recognition Challenge 2014.

The GoogLeNet team of Google researchers and interns including Liu, Christian Szegedy, Yangqing Jia, Pierre Sermanet, Scott Reed, Drago Anguelov, Dumitru Erhan and Andrew Rabinovich took first place in the areas of “object detection” and “object classification.”

Liu is a fourth-year doctoral student studying computer vision with advisers Alex Berg and Tamara Berg in UNC’s department of computer science. As an intern with Google this summer, Liu worked on improving object detection in images by “retraining a deep convolutional network” programmed by Google researchers.

Google researchers had previously trained the network to recognize whole images based on their content. Liu spent his summer applying machine-learning techniques to improve the network’s performance specifically for the task of detecting individual objects in an image.

The goal of the annual ImageNet Large Scale Visual Recognition Challenge is to allow research groups and, increasingly, start-ups and other companies to evaluate the accuracy of their latest algorithms for recognizing the content of images and detecting objects in photos. In addition to bragging rights for performance, seeing where various algorithms make mistakes helps researchers decide how to focus future research efforts.

The competition was organized by researchers at Stanford University and the University of North Carolina at Chapel Hill, with outside sponsorship from Facebook and Google. The competition attracted 38 entrants from 13 different countries.

Research in computer vision forms the basis for a variety of technologies, including image searches, video gaming and even automatic collision detection in automobiles. Improvements developed through benchmark challenges like the ImageNet contest can better equip modern-day ubiquitous computers to understand the world around them.

Read more from the computer science department.