Visual search is a new technology that’s likely to impact the mobile commerce space. It allows mobile users to snap pictures of just about anything on their devices and then search for relevant online content based on the photo itself.
This technology is useful for mobile users who are on the go and seeking certain qualities in a product without knowing or caring about the brand name. For example, someone with an iPhone could see a shirt in a store and want to search for shirts with similar aesthetics. One person might describe a pattern in a shirt differently than another.
Google is not the first to utilize mobile search, but it does provide Android users with access to the largest image database. Android users can use Google Goggles to match freshly-taken photos with Google’s massive image database. Google Goggles is a Google Labs application that provides mobile image search.
“Google Goggles works by attempting to match portions of an image submitted by a user against our corpus of images,” Google spokesperson Carolyn Penner said in an email. “When a match is found, we return search terms that are relevant to the matched image. When there is no match found, we show ‘similar images’ based on the texture and color from our image search database.”
Penner said Google collects these images during its web crawls and stores them in its image database. She stressed that the image used in the query would not be made public or crawled by Google without the user’s choosing.
“Just as he or she could for any picture, a user may choose to post the image used or share it publicly,” Penner said. “If the user shares the image to a public page that is crawlable, those images might turn up in a search at some point.”
Bing, Microsoft’s search engine, is testing its own visual search feature that creates image galleries and lets users scroll through product images. Instead of analyzing the image and using it to search, it relies on users to choose an image out of a gallery of hundreds of similar images. So far, it is not useable on mobile devices, and Bing would not comment on whether this would change anytime soon.
Other applications have made visual search available for users on other mobile devices, like iPhone. There are now several search-type apps in the iPhone App Store that make use of the device’s three-megapixel camera.
oMoby and Kooaba are both apps with similar functions to Google Goggles. iPhone users can snap pictures and search. The apps match the image up to others in their databases and then display shopping information. Both were released early this year, and both have gotten good reviews on the App Store.
Their strength seems to be in identifying media such as book covers, movie posters, DVD and CD covers, and video games. Kooaba’s image search is largely limited to these items, according to its site. oMoby can go further and recognize products with no text. On the site’s video demo, an oMoby user snapped a picture of a watch, and the app was able to identify the exact brand and model.
Like.com for Clothing and Accessories
Like.com, the visual search engine for clothing and apparel, recently launched an iPhone app, as well. It provides a visual shopping experience that allows customers searching for clothing, shoes and accessories to compare features based on the item’s aesthetics.
Participating merchants pay Like.com a per-click or per-action fee, according to a 2007 Practical eCommerce interview with a Like.com executive, Munjal Shah. Its users can search for clothing and apparel with similar colors, shapes, patterns or features.
Visual search could alter the search-marketing playing field. CNBC’s special report “Inside the Mind of Google” recognized this, saying that visual search will be critical for whoever wants to get the attention of millions of mobile users, who are projected to make up the majority of web traffic within a few years.