Tech
New Computer Program Teaches Itself Everything about Anything
Catherine Griffin
First Posted: Jun 14, 2014 05:51 AM EDT
Technology can make things easier, but it can also complicate your life. When you're trying to find something specific on the Internet, it can be difficult to sift through the large volume of information online. Now, scientists may have created a tool to help; they've made the first fully automated computer program that teaches everything to know about any visual concept.
The new program is called Learning Everything about Anything (LEVAN). It searches through millions of books and images on the Internet to learn all possible variations of a concept. Then, it displays the results to users as a comprehensive, browsable list of images.
"It is all about discovering associations between textual and visual data," said Ali Farhadi, one of the researchers, in a news release. "The program learns to tightly couple rich sets of phrases with pixels in images. This means that it can recognize instances of specific concepts when it sees them."
The program learns which terms are relevant by looking at the content to the images found on the Internet and then identifying characteristic patterns across them by using object recognition algorithms. More specifically, it draws on a rich set of phrases to understand and tag photos by their content and pixel arrangements, and not simply by the words displayed in captions. In all, users can browse through an existing library of about 175 concepts. If a concept doesn't exist, you can submit any search term and then program will begin generating a list of subcategory images that relate to the concept.
"Major information resources such as dictionaries and encyclopedias are moving toward the direction of showing users visual information because it is easier to comprehend and much faster to browse through concepts," said Santosh Divvala, one of the researchers, in a news release. "However, they have limited coverage as they are often manually curated. The new program needs no human supervision, and thus can automatically learn the visual knowledge for any concept."
LEVAN could be a new way to learn about any concept, in addition to a better way to search for something on the Internet.
See Now:
NASA's Juno Spacecraft's Rendezvous With Jupiter's Mammoth Cyclone
©2024 ScienceWorldReport.com All rights reserved. Do not reproduce without permission. The window to the world of science news.
More on SCIENCEwr
First Posted: Jun 14, 2014 05:51 AM EDT
Technology can make things easier, but it can also complicate your life. When you're trying to find something specific on the Internet, it can be difficult to sift through the large volume of information online. Now, scientists may have created a tool to help; they've made the first fully automated computer program that teaches everything to know about any visual concept.
The new program is called Learning Everything about Anything (LEVAN). It searches through millions of books and images on the Internet to learn all possible variations of a concept. Then, it displays the results to users as a comprehensive, browsable list of images.
"It is all about discovering associations between textual and visual data," said Ali Farhadi, one of the researchers, in a news release. "The program learns to tightly couple rich sets of phrases with pixels in images. This means that it can recognize instances of specific concepts when it sees them."
The program learns which terms are relevant by looking at the content to the images found on the Internet and then identifying characteristic patterns across them by using object recognition algorithms. More specifically, it draws on a rich set of phrases to understand and tag photos by their content and pixel arrangements, and not simply by the words displayed in captions. In all, users can browse through an existing library of about 175 concepts. If a concept doesn't exist, you can submit any search term and then program will begin generating a list of subcategory images that relate to the concept.
"Major information resources such as dictionaries and encyclopedias are moving toward the direction of showing users visual information because it is easier to comprehend and much faster to browse through concepts," said Santosh Divvala, one of the researchers, in a news release. "However, they have limited coverage as they are often manually curated. The new program needs no human supervision, and thus can automatically learn the visual knowledge for any concept."
LEVAN could be a new way to learn about any concept, in addition to a better way to search for something on the Internet.
See Now: NASA's Juno Spacecraft's Rendezvous With Jupiter's Mammoth Cyclone