A (digital) picture tells a thousand words
September 23, 2008 Researchers from the aceMedia project in Europe are now working on ways to generate searchable images that could see digital files come with content information, metadata and an intelligence layer that automatically generates word-searchable data about the image.
To achieve the searchable image, the project re-used, developed and combined a series of technologies that provide greatly enriched content information about an image. One such technology is software that can identify low-level visual descriptors, such as consistent areas of color that may be sky, sea, sand or possibly snow, and information about texture, edge and shape. By combining the low-level descriptors with sets of contextual rules held in domain ontologies (such as the fact that consistent areas of blue at the top of an image are likely to be sky, or that beach and snow are unlikely to appear in the same picture), the data can be converted into a rich information source.
Converting low-level descriptors into useful information poses greater challenges. To achieve this, data from low-level descriptors was also combined with the results from specific detectors, such as the kinds of face detectors that are commercially available on some cameras today. Users are also able to play their part, adding another layer of information in the form of rules that define their personal preferences, profiles and policies to create a personalized filing system. To enable easier searching, natural language processing techniques were also used, meaning you can use everyday language when searching for an image.
The researchers drew together their full range of technologies in an Autonomous Content Entity (ACE) Framework to demonstrate the benefits of automated content sharing and easier content management on a series of home network devices, including PCs, mobile phones and set-top boxes. Using the framework, ACEs can be created that contain all of the rules, metadata and content information so that they become a part of the image file. A scalable video codec has also been developed for adding searchable content to digital video clips.
The technologies developed in the EU-funded project have sparked interest from a range of commercial companies and, according to according to Yiannis Kompatsiaris, one of the lead researchers on aceMedia, could be in use within the next five years. And while the vision of the aceMedia project was to combine technologies, each delivering a piece to the overall information puzzle, they are not interdependent according to Kompatsiaris. “The tools we developed in aceMedia are scalable to many concepts and many environments,” he said.
Via: ICT Results.