Corbis, a provider of digital images and eVision, a leader in visual search technology, will begin supplying content for eVision’s proprietary eVe technology for its online visual search engine.
Under the terms of the agreement, Corbis and eVision will demonstrate the features and capabilities of the eVe visual search solution using a large selection of over 60,000 Corbis-branded images from their royalty-free collection.
launched its proprietary eVe technology at last month’s Seybold Seminars in San Francisco
eVe is designed to let users quickly find desired target products or brand assets using visual media and fewer clicks. Specifically, eVision’s technology increases search precision by enabling initiation and search for products based on desired visual features and “Visual Vocabulary.” The eVe technology integrates with existing search mechanisms and digital asset management systems.
Using eVe, developers can create applications that enable users to search using images and objects that contain visual similarity, rather than keywords and text. The software is built in Java, and works on Mac OS X, Solaris, Linux and Windows NT and 2000.
Currently, the most popular way of finding visual assets within Corbis’s Web sites is the use of text-based attributes, according to William Radcliffe, director of technology development at Corbis. However, many Corbis customers are more visually oriented and think in terms of colors, shapes and textures in addition to words, he said.
“The goal of this collaboration is to clearly demonstrate how this alternative to text search can significantly improve the user experience and allow Corbis users other options to find the content they need faster and more accurately than with text search alone,” Radcliffe said. “It is part of an ongoing effort to make the Corbis offerings the best they can possibly be. The eVe technology makes implementing this very easy.”
Canto, a supplier of digital asset management products and services, and eVision announced that the eVe technology has been fully integrated into Cumulus 5, Canto’s digital asset management program.
Through the integration of the eVe technology, Canto customers will be able to distill images into their representative visual features, based on object, color, texture and shape, company spokesperson Jason Taylor told MacCentral. The algorithms employed automatically segment each image into distinct object regions and generate scale-and rotation-independent descriptions of the regions, known as visual signatures.
These visual signatures are then organized into a proprietary indexing scheme for rapid and accurate retrievals. The algorithms can be applied to index image content, video content, audio content and any other digital patterns.