Researchers teach computers to search photos by subjectOctober 9th, 2008 - 4:29 pm ICT by IANS
Washington, Oct 9 (IANS) Penn State University researchers have developed a statistical approach, called ALIPR, that one day could make it easier to search the net for photographs.The public can participate in improving ALIPR or automatic linguistic indexing of pictures in real-time accuracy, by visiting www.alipr.com, uploading photographs, and evaluating whether the keywords that ALIPR uses to describe the photographs are appropriate.
ALIPR works by teaching computers to recognise the contents of photographs like buildings, people, or landscapes, rather than by searching for keywords in the surrounding text, as is done with most current image-retrieval systems.
They hope that eventually ALIPR can be used in industry for automatic tagging or as part of Internet search engines, said a release of Penn State.
“Our basic approach is to take a large number of photos - we started with 60,000 - and to manually tag them with a variety of keywords that describe their contents,” said Jia Li, an associate professor of statistics at Penn State.
“For example, we might select 100 photos of national parks and tag them with the following keywords: national park, landscape, and tree. We then would build a statistical model to teach the computer to recognize patterns in colour and texture among these 100 photos and to assign our keywords to new photos that seem to contain national parks, landscapes, and/or trees,” Li added.
Li, who developed ALIPR with her colleague James Wang, of Penn State, said that their approach appropriately assigns to photos at least one keyword among seven possible keywords about 90 percent of the time.
But, she added, the accuracy rate really depends on the evaluator. “It depends on how specific the evaluator expects the approach to be,” she said. “For example, ALIPR often distinguishes people from animals, but rarely distinguishes children from adults.”
- Computers will soon search photos on the web based on their contents - Oct 10, 2008
- Writing assignments boost students' critical thinking skills - Apr 02, 2011
- Updates on social networking sites worth more than $30m a day - Apr 13, 2011
- China's search engine market reaches $877 mn - Apr 28, 2012
- No more photos with Siberian tiger cubs in China - Jul 19, 2012
- For teens online tagged photos more precious than actual ones - May 10, 2011
- Now an iPhone app that translates languages - Jul 15, 2012
- 'Satyamev Jayate' grows more popular than Aamir! - May 14, 2012
- Smiling owl caught on camera in Washington Olympic National Park - Mar 01, 2012
- Santiniketan through a historic photographers' lens - Oct 11, 2011
- Why is state government against heritage tag for Western Ghats, question greens - Jul 09, 2012
- Great Himalayan National Park vies for World Heritage status - Jun 13, 2012
- Machines can't recognise images like humans as yet - Sep 10, 2009
- Monkeys recognise photos of their friends - Mar 18, 2011
- Teacher in topless photo scandal is back! - Oct 09, 2011