About
On the streets, you see someone wearing a T-Shirt you really like. You visit an online shopping site later that day, uploading an image of the shirt into the search bar and found an exact match. How is it possible that we don't need to search by words, but instead can use images now?  The "magic" behind this feature is a system for image querying. This guided project will help you build an image querying prototype. The project is split into two parts. The first part focuses on building the image encoding system and the second part focuses on building the image query system.

This application has many business use cases such as:

A Look at the Project Ahead
After completing this Guided Project, you will be able to:
- Setup an Image Encoding service that accepts input images and produces embeddings

- Explore techniques for generating embeddings

- Generate the embeddings for the dataset and save it on disc

- Import the embeddings dataset from previous notebook

- Generate embeddings for a query image

- Search the embeddings dataset for closest match for the given query image

What You'll Need
This course mainly uses Python. Prior experience with libraries such as Tensorflow and PIL are recommended but not required.  System requirements for if you run the labs on a local machine include TensorFlow 2.x and 2GB of storage (for storing the dataset).

Frequently Asked Questions

> Do I need to install any software to participate in this project?

Everything you need to complete this project will be provided to you via the Skills Network Labs and it will all be available via a standard web browser.

> What web browser should I use?

The Skills Network Labs platform works best with current versions of Chrome, Edge, Firefox, Internet Explorer, or Safari.

Your Instructors
Kai Niu