eBay Launches New Image Search Feature

eBay has launched a new image search feature for its mobile app that enables users to simply drag and drop images into the search bar.

“When we introduced Image Search last year, we used computer vision to make it easier to find like items from the world around you in our global marketplace. Now, you can explore our marketplace further with just a swipe of your finger,” wrote Seema Jethani, manager of product management and search platforms at eBay. “As we continue to make the entire internet shoppable using computer vision, we’re evolving the search experience on eBay to help you more easily find what you want. Starting in August, our new visual shopping feature [will allow] you to drag an image to the search bar to find more items that look like the image that you love.”

For example, she noted, if a user types “egg chair” into the eBay search bar, he might see a list of results. If the user prefers one particular chair, he can now search for products that look like it by dragging and dropping the image of the chair into the search bar to see visually similar results.

The new shopping feature is powered by artificial intelligence, and utilizes deep learning networks, known as convolutional neural networks, to process the images.

“Behind the scenes, when you submit your image in the search field, the neural network converts your image into a vector representation,” Jethani added.“Then, the vector representation of the image that you submit is compared against more than 1.1 billion live listings in eBay’s marketplace, using nearest neighbor search. Finally, we surface the best-matched items, ranked by visual similarity.”

The visual shopping feature will be available on Android and iOS for the U.S., the U.K., Germany and Australia.


New PYMNTS Report: Preventing Financial Crimes Playbook – July 2020 

Call it the great tug-of-war. Fraudsters are teaming up to form elaborate rings that work in sync to launch account takeovers. Chris Tremont, EVP at Radius Bank, tells PYMNTS that financial institutions (FIs) can beat such highly organized fraudsters at their own game. In the July 2020 Preventing Financial Crimes Playbook, Tremont lays out how.