Microsoft Cognitive Services for Developers: 1 Vision

Learn how to use Microsoft Cognitive Services to embed AI in your applications. This course covers the vision APIs for text, image, and facial recognition.

Microsoft Cognitive Services opens up artificial intelligence to every developer. Cognitive Services gives us access to powerful AI algorithms that allow our apps to hear, speak, identify, and interpret information using natural methods of communication. This course explores the vision APIs, which are responsible for text and image recognition and facial detection. Instructor Sahil Malik shows you how to configure Cognitive Services and use the vision capabilities to describe, tag, and categorize images; identify clip art vs. line art; moderate content; "read" text and handwriting; and recognize facial characteristics such as age, gender, emotion, hair color and type, and more. Sahil also shows how to build a facial recognition program that can recognize people in images by comparing them to a library of known persons.

Introduction

• Welcome
• What you should know
• Using the exercise files

1. The Basics

• Introduction
• Set up the Node.js project
• Add TypeScript debugging support
• Add helper code and demonstrate debugging

2. Describing, Categorizing, and Tagging

• Provision Vision API in Azure and configure project
• Write code to describe an image using the Vision API
• Run the code to describe an image using the Vision API
• Returning multiple descriptions
• Defining the analysis business object
• Writing the analyze method
• Using the analyze method to describe an image
• Tags and categories

3. Performing Deeper Analysis

• Analyzing colors
• Identifying line drawing and clip art vs. real pictures
• Identifying adult vs. racy content
• Generating thumbnails

4. Working with Models

• Recognizing celebrities
• Recognizing landmarks

5. Handwriting and OCR

• Set up the Node.js project
• Submit an image for handwriting or OCR
• Polling for status of submitted image
• Synchronous treatment for OCR
• Performing OCR
• Performing handwriting recognition

6. Working with Faces

• Recognize faces using the Vision API
• Provision the Emotion API and update the project
• Making code changes to analyze emotions
• Recognizing emotions for multiple pictures
• Provision the Face API and update the project

7. Identifying People

• Recognizing faces: The theory
• Set up your project
• Set up your sample dataset
• Read the dataset
• Organize your code and explain what you need to do
• Create a person group
• Delete or train the person group
• Create a person and add a face
• Add detection and identification functions
• Create a person group using helper classes
• Train a person group
• Detecting and identifying a person
• Run code and delete the person group

Conclusion

• Next steps