Computers will soon search photos on the web based on their contents
Washington, October 10 : A pair of Penn State researchers has come up with a statistical approach that may one day make Internet searches for photographs quite easier.
The tool they use for this purpose is called Automatic Linguistic Indexing of Pictures in Real-Time (ALIPR), which works by teaching computers to recognize the contents of photographs, such as buildings, people, or landscapes.
The procedure is very different from searching for keywords in the surrounding text, as is done with most current image-retrieval systems.
The team recently received a patent for an earlier version of the approach called ALIP, and is in the process of obtaining another patent for the more sophisticated ALIPR.
According to the researchers, people can help them improve ALIPR''s accuracy by visiting a designated Web site (http://www.alipr.com), uploading photographs, and evaluating whether the keywords that it uses to describe the photographs are appropriate.
They hope that eventually ALIPR can be used in industry for automatic tagging or as part of Internet search engines.
"Our basic approach is to take a large number of photos -- we started with 60,000 photos -- and to manually tag them with a variety of keywords that describe their contents. For example, we might select 100 photos of national parks and tag them with the following keywords: national park, landscape, and tree," said Jia Li, an associate professor of statistics at Penn State.
"We then would build a statistical model to teach the computer to recognize patterns in color and texture among these 100 photos and to assign our keywords to new photos that seem to contain national parks, landscapes, and/or trees. Eventually, we hope to reverse the process so that a person can use the keywords to search the Web for relevant images," she added.
She and her colleague James Wang, a Penn State associate professor of information sciences and technology, said that their approach appropriately assigns to photos at least one keyword among seven possible keywords about 90 percent of the time.
She, however, added that the accuracy rate really depended on the evaluator.
"It depends on how specific the evaluator expects the approach to be. For example, ALIPR often distinguishes people from animals, but rarely distinguishes children from adults," she said.
Although the team wants to improve ALIPR''s accuracy, Li said that she did not believe the approach ever will be 100-percent accurate.
"There are so many images out there and so many variations on the images'' contents that I don''t think it will be possible for ALIPR to be 100-percent accurate," she said.
"ALIPR works by recognizing patterns in color and texture. For example, if a cat in a photo is wearing a red coat, the red coat may lead ALIPR to tag the photo with words that are irrelevant to the cat. There is just too much variability out there," she added.
She is presently pursuing some new ideas that might help her achieve better recognition of image semantics. (ANI)