What this book covers
Chapter 1, Computer Vision Applications and AWS AI/ML Overview, provides an introduction to CV and summarizes use cases where CV can be applied to solve business challenges. It also includes an overview of the AWS AI/ML services.
Chapter 2, Interacting with Amazon Rekognition, covers an overview of Amazon Rekognition and details the different capabilities available, including walking through the Amazon Rekognition console, and how to use the APIs.
Chapter 3, Creating Custom Models with Amazon Rekognition Custom Labels, provides a detailed introduction to Amazon Rekognition Custom Labels, what its benefits are, and a code example to train a custom object detection model.
Chapter 4, Using Identity Verification to Build a Contactless Hotel Check-In System, dives deep into a real-world use case using Amazon Rekognition and other AWS AI services to build applications that demonstrate how to solve business challenges using core CV capabilities. A code example is provided to build a mobile application for customers to register their faces and check into a fictional hotel kiosk system.
Chapter 5, Automating a Video Analysis Pipeline, dives deep into a real-world use case using Amazon Rekognition to build an application that demonstrates how to solve business challenges using core CV capabilities. A code example is provided to build a real-time video analysis pipeline using Amazon Rekognition Video APIs.
Chapter 6, Moderating Content with AWS AI Services, dives deep into a real-world use case using Amazon Rekognition and other AWS AI services to build applications that demonstrate how to solve business challenges using core CV capabilities. A code example is provided to build content moderation workflows.
Chapter 7, Introducing Amazon Lookout for Computer Vision, provides a detailed introduction to Amazon Lookout for Vision, what its functions are, and a code example to train a model to detect anomalies.
Chapter 8, Detecting Manufacturing Defects Using CV at the Edge, dives deeper into Amazon Lookout for Vision, covers the benefits of deploying CV at the edge, and walks through a code example to train a model to detect anomalies in manufacturing parts.
Chapter 9, Labeling Data with Amazon SageMaker Ground Truth, provides a detailed introduction to Amazon SageMaker Ground Truth, what its benefits are, and a code example to integrate a human labeling job into offline data labeling workflows.
Chapter 10, Using Amazon SageMaker for ComputerVision, dives deeper into Amazon SageMaker, covers its capabilities, and walks through a code example to train a model using a built-in image classifier.
Chapter 11, Integrating Human-in-the-Loop with Amazon Augmented AI, provides a detailed introduction to Amazon Augmented AI (Amazon A2I), what its functions are, and a code example that uses human reviewers to improve the accuracy of your CV workflows.
Chapter 12, Best Practices for Designing an End-to-End CV Pipeline, covers best practices that can be applied to CV workloads across the entire ML lifecycle, including considerations for cost optimization, scaling, security, and developing an MLOps strategy.
Chapter 13, Applying AI Governance in CV, discusses the purpose of establishing an AI governance framework, introduces Amazon SageMaker for ML governance, and provides an overview of the importance of mitigating bias.