AI Case Study

Microsoft's Bing enhances image search uses deep learning to automatically detect objects in an image and give more comprehensive search results

Microsoft Bing's image search was constrained by the fact that it did not automatically detect objects in an image but users had to mark the objects they want to search manually. Using deep learning Bing can automatically identify several objects in an image and associate key words to look for offering better search results.

Industry

Technology

Internet Services Consumer

Project Overview

"The Bing mobile application will now search the web using images, and uses artificial intelligence (AI) to give more comprehensive search results. Bing uses AI to determine the object in the picture and gather information.

Users can shop fashion or home furnishings with the app. Bing will identify the object and find similar items on the web or stores."

Reported Results

Results undisclosed; However Bing's image search is more efficient and easier to use

Technology

Object Detection Model

"One crucial characteristic shared by most object detection algorithms is generation of category-independent region hypotheses for recognition, or “region proposals”. As compared to other frameworks where region proposals are generated offline, Faster R-CNN speeds the process up significantly enough for it to be done online. It achieves this by creatively sharing the full-image convolutional features between Region Proposal Network (RPN) and the detection network."

Function

R And D

Product Development

Background

Bing's image search was not readily usable for customers as they had to manually mark the objects to search.

Benefits

Data