Through all the technological advancements that have come with the internet, text box searches have remained consistent. Everyone is used to the way we search things, type words into a rectangular box to get the search results you are wanting, but recent advances in image recognition software may offer some changes to this. Major online names like Pinterest and Shoes.com are testing new ways to preform searches by using images.
Both these companies have embraced a method called deep learning which allows software to match human features for image recognition. This is the method that is used in Google’s image search and the new photo organization they launched this summer called Automated Helpers.
The visual search allows users to draw a box around an image then place it in a search box to find results that are similar. Some of these search results come equipped with buy buttons making searching and purchasing so much simpler. Amazon included an app that finds products similar to photos taken by users on their Fire smartphone they released last year.
Kevin Jing, the head of visual search at Pinterest, feels that this visual searching method will soon become indispensable. He feels this method of searching items is much more accurate and each year offers many improvements to the software.
Shoes.com is testing a different method of the search by making use of an image-processing technology aimed at retailers that was developed by AI startup company Sentient. This technology is currently being used in their women’s boots section of their website. So far the CEO of Shoes.com says that he has seen evidence that the visual filter is actually increasing sales and he is looking to expand the search feature into other departments.
One company leading the market in visual search technology is Slyce. Their technology allows users to drop photos and pictures in the search box and receive results that are fitted to the original search image. This technology is available in an app for smartphones and other devices.
The technology they employ allows their search engine to instantly recognize objects then use that knowledge to recognize and connect consumers with similar items available at online retailers. The technology is so advanced users can even insert photos of any items they see and like in the real world into a search box to find exactly what they are shopping for.