Chinese online consumers now can purchase products directly from the British high-end fashion brand Burberry. Burberry Group PLC launched its online store Tuesday on ...
Image recognition technology searches databases by picture instead of text
France-based LTU Technologies has launched in the U.S. this week to pursue e-retail and other markets with a search software that understands how images are created and seeks other similar images.
LTU Technologies, a French technology developer that specializes in visual content management, this week launched U.S. operations to offer solutions including visually-based database search. While its patented image recognition technology also has application in content filtering, indexing and digital rights management, e-retail is among the key target markets in the U.S., according to the company.
The image recognition technology has been live on Ibazar, a French online auction company recently acquired by eBay, where 6% of site visitors used the tool, says Jody Webster, general manager of the company’s U.S. subsidiary. LTU’s Image-Seeker technology recognizes images as a set of numbers that describe pixels, and matches the description to similar images in a database. The technology opens the door to “find this” features that would let online shoppers search a database for images of pants, for example, that are similar to a product image the user has selected. LTU’s technology can also allow users to scan in or upload digital images to search a database for similar items.
The technology unlocks new possibilities for cross-selling and upselling merchandise online, Webster says, increasing the utility of visual content by making it searchable and retrievable. “Many of the content management tools in use today are text-based and overlook the essential need to manage content by understanding the images themselves,” she adds.