Visual search is a method of displaying similar images to one uploaded by a user to a retail website. Similar images are found by transforming an image into a feature vector using a CNN. Visual search has a lot of applications in online shopping as it compliments textual search for a better and more refined way of expressing a user's choice of product. Shoppers like visual discovery and find it something unique that is not available in a traditional shopping experience.
In this chapter, we will use the concepts of deep neural networks learned in Chapter 4, Deep Learning on Images, and Chapter 5, Neural Network Architecture and Models. We will use transfer learning to develop a neural network model for our image classes and apply it for visual search. The exercises in this chapter will help you to develop sufficient practical knowledge...
