Archive for November 20th, 2012

ImageNet

November 20, 2012

The Internet provides us with a mountain of digital images from which you can find just about anything you are looking for.  Sometimes, but not always. Why? Because an image can be found on the Internet only if the text entered by a searcher matches the text used to label it.  To help organize this heap of digital confusion, computer scientists at Stanford and Princeton are working on the world’s largest visual database that mimics the human vision system. The database called ImageNet is organized around hierarchical categories called the WordNet. Each category (node) is represented by hundreds or thousands of images — on the average over 500 images per node. ImageNet utilizes Mechanical Turk workers to identify and categorize images. At the moment, ImageNet has over 14,000,000 images indexed into nearly 22,000 categories.
For more on ImageNet, read “Seeking a Better Way to Find Web Images” (New York Times, November 19, 2012)

Conservators come to the aid of Eyebeam

November 20, 2012

Like many Chelsea art galleries and art agencies, Eyebeam was devastated by hurricane Sandy.  Eyebeam, a nonprofit focusing on new media, was under three feet of dirty, toxic saltwater and its AV equipment, media and computers, including its archives documenting years of exhibitions and non-traditional media art projects, were extremely vulnerable and exposed.  New York City conservators mobilized and came to the nonprofits aid.  Watch a video produced by documentary artist Jonathan Minard capturing the conservation efforts of this community.
To contribute to Eyebeam’s recovery efforts, go here.