Collegian Venues - your weekend starts here
  Child Development and Family Council

Get a deal with Daily Collegian Coupon Corner
  The Daily Collegian Online	 - Published independently by students at Penn State SCIHEALTH
[ Tuesday, Nov. 7, 2006 ]

Intelligent computer describes pictures

Collegian Staff Writer

Just as "Google it" has become an everyday phrase, two Penn State researchers hope to make the words "ALIPR it" just as common.

Jia Li, associate professor of statistics, and James Wang, associate professor of information science and technology, have created a program called Automatic Linguistic Indexing of Pictures-Real Time (ALIPR).

"It is a system that if you give it a picture without any textual information, just like a jpeg picture, the computer will automatically describe what the picture is about," Li said.

More than a billion pictures or images on search engines do not have text associated with the image, Wang said.

"If there is no text with the picture, then you can't match keywords to the picture, and the picture is basically invisible to search engines," Li said.

ALIPR will allow more images to be located by teaching the computer keywords associated with an image, similar to teaching a human what a tiger looks like, Wang said.

"When using ALIPR, a computer can suggest keywords for uploaded

pictures," Wang said. "These computers can have 15 suggestions for keywords and a person can add more so that every time the words are searched, the image will be located."

Many images are not annotated because it is very labor-intensive, Wang added.

"We have made a lot of progress over the years, but not as successful as text," Wang said. "In text search you don't have to understand the words, but when searching for images, we believe you really need computers to understand the images."

Mor Naaman, a scientist at Yahoo research Berkeley, said he believes the computers will not be able to analyze the difference between context and content.

"This system is a good start, but it will never be complete or perfect," he said.

Naaman explained the difficulty he thinks a computer will have distinguishing between a picture's content and context using an example of a clock tower.

"The computer will believe an image of a clock tower at night and an image of it in the morning are two different pictures because it will not understand context of night and day," he said.

PHOTO: ddd

Naaman said he does believe this program will enable people to search easier, but there are still flaws.

"The computer will eventually draw some conclusions about pictures on its own and not need to be annotated," Wang said. "That is our goal -- to try to use a large amount of images to teach the computers to recognize certain keywords that are not necessarily linked to annotated pictures."

Anyone can access the Web site at ALIPR.com, he added.

"Hopefully by spreading the news more people will come help us," Wang said. "Anyone in the world can teach a computer using ALIPR."

Even students at Penn State can upload any picture on ALIPR that ends with a URL, Wang said. The computer will suggest 15 keywords and students can verify the keywords accurately describe the image, he added.

"This will help the training and teaching process of the computer," Wang said.

ALIPR was recently presented at the Association for Computing Machinery conference in Santa Barbara, Calif. It is the most prestigious and largest multimedia community in the world about computer science, Wang said.

"People were really interested because we showed a demo live, and this is the only system that works for hundreds of English words," Wang said.

Li said she believes it is a great stride in searching for images, too.

"The technology itself is very ambitious," Li said. "If a computer can actually understand what the image is about, there are many potential uses."

Wang said there are other implications with this program, such as satellite pictures and medical uses.

"Just like humans, computers can be trained," Wang said. "Soon hopefully the computers can learn about art pictures and then will recognize Van Gogh paintings."

In the next few years, Wang said they also hope to break the language barrier, allowing the public to search for images and pictures in different languages.

"It is a very powerful ability," Li said.


 

Send an Opinion Letter to the Editor about this article.


Print this pagePrint this page   E-mail this pageE-mail this page   





TOP  HOME
Copyright 2006 Collegian Inc.
Updated: 2006-11-6  21:52:00   -4
Requested: 2006-11-7  12:36:04   -4
Created: 2006-11-6  21:51:09   -4