book

Computer Vision on AWS

Name: Computer Vision on AWS
ISBN: 9781801078689

by Lauren Mullennex, Nate Bachmeier, Jay Rao

March 2023

Intermediate to advanced

324 pages

6h 44m

English

Packt Publishing

Read now

Unlock full access

Computer Vision on AWS
ContributorsAbout the authorsAbout the reviewer
Preface
Who this book is forWhat this book coversTo get the most out of this bookDownload the example code filesConventions usedGet in touchShare Your ThoughtsDownload a free PDF copy of this book
Part 1: Introduction to CV on AWS and Amazon Rekognition
Chapter 1: Computer Vision Applications and AWS AI/ML Services Overview
Technical requirementsUnderstanding CVCV architecture and applicationsData processing and feature engineeringData labelingSolving business challenges with CVContactless check-in and checkoutVideo analysisContent moderationCV at the edgeExploring AWS AI/ML servicesAWS AI servicesAmazon SageMakerSetting up your AWS environmentCreating an Amazon SageMaker Jupyter notebook instanceSummary
Chapter 2: Interacting with Amazon Rekognition
Technical requirementsThe Amazon Rekognition consoleUsing the Label detection demoExamining the API requestExamining the API responseOther demosMonitoring Amazon RekognitionQuick recapDetecting Labels using the APIUploading the images to S3Initializing the boto3 clientDetect the LabelsUsing the Label informationUsing bounding boxesQuick recapCleanupSummary
Chapter 3: Creating Custom Models with Amazon Rekognition Custom Labels
Technical requirementsIntroducing Amazon Rekognition Custom LabelsBenefits of Amazon Rekognition Custom LabelsCreating a model using Rekognition Custom LabelsDeciding the model type based on your business goalCreating a modelImproving the modelStarting your modelAnalyzing an imageStopping your modelBuilding a model to identify Packt’s logoStep 1 – Collecting your imagesStep 2 – Creating a projectStep 3 – Creating training and test datasetsStep 4 – Adding labels to the projectStep 5 – Drawing bounding boxes on your training and test datasetsStep 6 – Training your modelValidating that the model worksStep 1 – Starting your modelStep 2 – Analyzing an image with your modelStep 3 – Stopping your modelSummary
Part 2: Applying CV to Real-World Use Cases
Chapter 4: Using Identity Verification to Build a Contactless Hotel Check-In System
Technical requirementsPrerequisitesCreating the image bucketUploading the sample imagesCreating the profile tableIntroducing collectionsCreating a collectionDescribing a collectionDeleting a collectionQuick recapDescribing the user journeysRegistering a new userAuthenticating a userRegistering a new user with an ID cardUpdating the user profileImplementing the solutionChecking image qualityIndexing face informationSearch existing facesQuick recapSupporting ID cardsReading an ID cardUsing the CompareFaces APIQuick recapGuidance for identity verification on AWSSolution overviewDeployment processCleanupSummary
Chapter 5: Automating a Video Analysis Pipeline
Technical requirementsCreating the video bucketUploading content to Amazon S3Creating the person-tracking topicSubscribing a message queue to the person-tracking topicCreating the person-tracking publishing roleSetting up IP camerasQuick recapUsing IP camerasInstalling OpenCVInstalling additional modulesConnecting with OpenCVViewing the frameUploading the frameReporting frame metricsQuick recapUsing the PersonTracking APIUploading the video to Amazon S3Using the StartPersonTracking APIReceiving the completion notificationUsing the GetPersonTracking APIReviewing the GetPersonTracking responseViewing the frameQuick recapSummary
Chapter 6: Moderating Content with AWS AI Services
Technical requirementsModerating imagesUsing the DetectModerationLabels APIUsing top-level categoriesUsing secondary-level categoriesPutting it togetherQuick recapModerating videosCreating the supporting resourcesFinding the resource ARNsUploading the sample video to Amazon S3Using the StartContentModeration APIExamining the completion notificationUsing the GetContentModeration APIQuick recapUsing AWS Lambda to automate the workflowImplement the Start Analysis HandlerImplementing the Get Results HandlerPublishing function changesExperiment with the end-to-endSummary

Part 3: CV at the edge
Chapter 7: Introducing Amazon Lookout for Vision
Technical requirementsIntroducing Amazon Lookout for VisionThe benefits of Amazon Lookout for VisionCreating a model using Amazon Lookout for VisionChoosing the model type based on your business goalsCreating a modelStarting your modelAnalyzing an imageStopping your modelBuilding a model to identify damaged pillsStep 1 – collecting your imagesStep 2 – creating a projectStep 3 – creating the training and test datasetsStep 4 – verifying the datasetStep 5 – training your modelValidating it worksStep 1 – trial detectionStep 2 – starting your modelStep 3 – analyzing an image with your modelStep 4 – stopping your modelSummary
Chapter 8: Detecting Manufacturing Defects Using CV at the Edge
Technical requirementsUnderstanding ML at the edgeDeploying a model at the edge using Lookout for Vision and AWS IoT GreengrassStep 1 – Launch an Amazon EC2 instanceStep 2 – Create an IAM role and attach it to an EC2 instanceStep 3 – Install AWS IoT Greengrass V2Step 4 – Upload training and test datasets to S3Step 5 – Create a projectStep 6 – Create training and test datasetsStep 7 – Train the modelStep 8 – Package the modelStep 9 – Configure IoT Greengrass IAM permissionsStep 10 – Deploy the modelStep 11 – Run inference on the modelStep 12 – Clean up resourcesSummary
Part 4: Building CV Solutions with Amazon SageMaker
Chapter 9: Labeling Data with Amazon SageMaker Ground Truth
Technical requirementsIntroducing Amazon SageMaker Ground TruthBenefits of Amazon SageMaker Ground TruthAutomated data labelingLabeling Packt logos in images using Amazon SageMaker Ground TruthStep 1 – collect your imagesStep 2 – create a labeling jobStep 3 – specify the job detailsStep 4 – specify worker detailsStep 5 – providing labeling instructionsStep 6 – start labelingStep 7 – output dataImporting the labeled data with Rekognition Custom LabelsStep 1 – create the projectStep 2 – create training and test datasetsStep 3 – model trainingSummary
Chapter 10: Using Amazon SageMaker for Computer Vision
Technical requirementsFetching the LabelMe-12 datasetInstalling TensorFlow 2.0Installing matplotlibUsing the built-in image classifierUpload the dataset to Amazon S3Prepare the job channelsStart the training jobMonitoring and troubleshootingQuick recapHandling binary metadata filesDeclaring the Label classReading the annotations fileDeclaring the Annotation classValidate parsing the fileRestructure the filesLoad the datasetQuick recapSummary
Part 5: Best Practices for Production-Ready CV Workloads
Chapter 11: Integrating Human-in-the-Loop with Amazon Augmented AI (A2I)
Technical requirementsIntroducing Amazon A2ICore concepts of Amazon A2ILearning how to build a human review workflowCreating a labeling workforceSetting up an A2I human review workflow or flow definitionInitiating a human loopLeveraging Amazon A2I with Amazon Rekognition to review imagesStep 1 – Collecting your imagesStep 2 – Creating a work teamStep 3 – Creating a human review workflowStep 4 – Starting a human loopStep 5 – Checking the human loop statusStep 6 – Reviewing the output dataSummary
Chapter 12: Best Practices for Designing an End-to-End CV Pipeline
Defining a problem that CV can solve and processing dataDeveloping a CV modelTrainingEvaluatingTuning Deploying and monitoring a CV modelShadow testingA/B testingBlue/Green deployment strategyMonitoringDeveloping an MLOps strategy SageMaker MLOps featuresWorkflow automation toolsUsing the AWS Well-Architected FrameworkCost optimizationOperational excellenceReliabilityPerformance efficiencySecuritySustainabilitySummary
Chapter 13: Applying AI Governance in CV
Understanding AI governanceDefining risks, documentation, and complianceData risks and detecting biasAuditing, traceability, and versioningMonitoring and visibilityMLOpsResponsibilities of business stakeholdersApplying AI governance in CVTypes of biasesMitigating bias in identity verification workflowsUsing Amazon SageMaker for governanceML governance capabilities with Amazon SageMakerAmazon SageMaker Clarify for explainable AISummary
Index
Why subscribe?
Other Books You May EnjoyPackt is searching for authors like youShare Your ThoughtsDownload a free PDF copy of this book

Overview

Unlock the power of computer vision (CV) using AWS with this actionable guide. Learn to design, implement, and deploy end-to-end CV pipelines efficiently utilizing AWS services such as Amazon Rekognition and Amazon SageMaker. By mastering these tools and best practices, you'll be equipped to enhance business processes and automation through intelligent visual data processing.

What this Book will help me do

Gain expertise in building scalable computer vision solutions on AWS
Learn to implement bias mitigation and ethical AI practices in CV pipelines
Master the usage of Amazon Rekognition and Amazon Lookout for Vision
Automate CV model training and deployment using Amazon SageMaker
Apply practical techniques for cost-effective CV model management

Author(s)

Lauren Mullennex, Nate Bachmeier, and Jay Rao are seasoned experts in AI and cloud technology with extensive practical experience in designing and deploying AWS-based solutions. With a collective passion for teaching, they have crafted this book to provide deep insights and hands-on techniques for building computer vision systems, drawing from real-world projects.

Who is it for?

Ideal for machine learning engineers and data scientists aiming to integrate computer vision capabilities into their workflows using cloud technologies. This book assumes foundational knowledge of AWS services and proficiency in Python programming to maximize the learning experience.

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781801078689

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills