News
Researchers teach computers to keyword images
Researchers at Penn State University are claiming to have made a major breakthrough that could impact search engines that rely on keywording or tagging to identify images. The researchers have "taught" computers how to interpret images using a vocabulary of up to 330 English words, so that a computer can describe a photograph of two polo players, for instance, as "sport," "people," "horse," "polo."
The potential for this new system is that image libraries and photo sharing communities may be able to further automate their keywording processes, resulting in an efficient standardized method to keyword images. In addition to saving time, the cost savings for even a small stock photo library could be tremendous.
The ALIPR system (Automatic Linguistic Indexing of Pictures-Real Time) was developed in order to apply text tags to images as they are uploaded to the internet making images visible to Web users. The technology co-developed by James Wang, associate professor in the Penn State College of Information Sciences and Technology is described in a paper, "Real-Time Computerized Annotation of Pictures," given at the recent ACM Multimedia 2006 conference in Santa Barbara, Calif., and authored by Jia Li, associate professor, Department of Statistics, and Wang. Penn State has filed a provisional patent application on the invention.
"By inputting tens of thousands of images, we have trained computers to recognize certain objects and concepts and automatically annotate those new or unseen images," Wang said. "More than half the time, the computer's first tag out of the top 15 tags is correct."
In addition, for 98 percent of images tested, the system has provided at least one correct annotation in the top 15 selected words. The system, which completes the annotation in about 1.4 seconds, also can be applied to other domains such as art collections, satellite imaging and pathology slides, Wang said.
A demonstration of the ALIPR system can be found at www.alipr.com online. Research on the ALIPR system was supported by the National Science Foundation.
----
Related Stories:
Meditations on keywording (February 9, 2006)
Digital Railroad announces partnership with JaincoTech (November 2, 2006)
OnAsia Digital keywording Britain's National Trust Photo Library (October 16, 2006)
Alamy seeking patent for new search technology (July 14, 2006)
Posted in: Keywording / Metadata, News, Photographers, Stock Art Buyers
