PRZOOM - /newswire/ -
Palo Alto, CA, United States, 2012/07/11 - Superfish, Inc.'s ShapeRank examines images’ visual features and then does something no other visual search technology does - it examines the geometric relationships between them, to present images with visually similar content more accurately.
Superfish, Inc., a global leader in visual search technology, today announced one of its finest technological developments with ShapeRank. ShapeRank examines images’ visual features and then does something no other visual search technology does - it examines the geometric relationships between them, to present images with visually similar content more accurately. This additional dimension of geometry improves visual search’ precision and performance by several orders of magnitude and allows Superfish to deliver precise results.
While the use of geometry in visual search is a relatively intuitive concept, Superfish is the only company doing this at Internet scale. Other visual search companies and technologies out there can match exact images, but Superfish specializes in the ability to identify and rank results by similarity as well as matching near identical images. No one else has this capability.
“Visual search is currently in its early stages, but we are defining the way everyone will be searching everything in the near future. Photo sharing companies like Facebook, Pinterest and Twitter are using more images than ever before,” said Adi Pinhas, Co-Founder of Superfish. “Our vision is that visual search will allow people to use any picture as a search query and get instant content to their search queries in real time. ShapeRank is yet another important step in that direction.”
While text-based search engines are useful in many ways, they are unable to search the exponentially growing number of untagged images on the web, not to mention the number of images people simply cannot describe in words in a text-based search engine. Superfish’s core technology uses images (as opposed to keywords) as search queries for products and an algorithmic real-time visual process allowing for massive scale and coverage.
Current technologies employ categorization, object or scene recognition, or near-duplicate image matching techniques in order to perform a visual search. While these approaches can be effective in narrow use cases or for detecting exact image matches, they cannot detect and rank visually similar images. ShapeRank allows users to go beyond exact image matching and identify images that are visually similar and sort them by relevance in their web searches. The closer to “near identical” these images are, the higher they rank in search priority, and the faster people will find what they are seeking online.
Superfish provides a visual shopping platform called WindowShopper, demonstrating the power of visual search for shopping. With hundreds of developers already using this platform, Superfish has served more than 20 million users and nearly 1 billion searches to date. With more than 200 million products available from 50,000+ online stores, users experience visual search with the most comprehensive shopping solution. Our product proves that the technology works and can generate instant revenue for partners and distributors.
Superfish (Superfish.com) is a global leader in visual search technology. Founded in 2006 with the vision of enabling searches to be conducted by using images where text search isn’t possible or efficient. Superfish has developed advanced visual search technology that eliminates the frustration of searching for images without knowing exactly what an item is called or how to describe it in a typical text-based search engine. The Superfish visual search engine is the first and only technology to analyze an image 100% algorithmically, providing similar and near identical images in real time without the need for text tags or human intervention. Superfish provides a visual shopping platform called WindowShopper – a free browser add-on that demonstrates the power of visual search for shopping.