© 2016, Amazon Web Services, Inc. or its Affiliates. All rights reserved.

David Pearson

December 13, 2016

Amazon RekognitionEasy-to-Use Deep Learning-based Image

Recognition

Images – universal, ubiquitous, & essential

Images – explosive growth trends

Source: InfoTrends Worldwide Consumer Photos Captured and Stored.

2013 -2017 prepared for Mylio.

Amazon S3 – Images Stored

Images – explosive growth trends

Amazon Rekognition

Deep learning-based image recognition service

Search, verify, and organize millions of images

Object and Scene

DetectionFacial

Analysis

Face

Comparison

Facial

Recognition

DetectLabels

Object and Scene DetectionDetect objects, scenes, and concepts in images

Amazon Rekognition API

Amazon Rekognition API

DetectLabels

{

"Confidence": 94.62968444824219,

"Name": "adventure"

},

{

"Confidence": 94.62968444824219,

"Name": "boat"

},

{

"Confidence": 94.62968444824219,

"Name": "rafting"

},

. . .

Flower

ChairCoffee Table

Living Room

Indoors

Object and Scene Detection

Maple

Villa

Plant

Garden

Water

Swimming Pool

Tree

Potted Plant

Backyard

Object and Scene Detection

Using Rekognition Object and Scene Detection

Photo-sharing apps can power smart searches

and quickly find cherished memories, such as

weddings, hiking, or sunsets.

Vacation rental markets can automatically label

host-uploaded images with tags, such as

fireplace, kitchen, or swimming pool.

Travel sites and forums can classify user

generated images with labels such as beach,

camping, or mountains.

Object and Scene Detection – Use CaseDynamic Search Indexing

AMAZON ELASTICSEARCH

AMAZON S3

Amazon Rekognition API

Facial AnalysisDetect face and key facial characteristics

DetectFaces

Amazon Rekognition API

DetectFaces

[

{

"BoundingBox": {

"Height": 0.3449999988079071,

"Left": 0.09666666388511658,

"Top": 0.27166667580604553,

"Width": 0.23000000417232513

},

"Confidence": 100,

"Emotions": [

{"Confidence": 99.1335220336914,

"Type": "HAPPY" },

{"Confidence": 3.3275485038757324,

"Type": "CALM"},

{"Confidence": 0.31517744064331055,

"Type": "SAD"}

],

"Eyeglasses": {"Confidence": 99.8050537109375,

"Value": false},

"EyesOpen": {Confidence": 99.99979400634766,

"Value": true},

"Gender": {"Confidence": 100,

"Value": "Female”}

Demographic Data

Facial Landmarks

Sentiment Expressed

Facial Analysis

General AttributesImage Quality

Demographic Data

Facial Landmarks

Sentiment Expressed

General Attributes

Facial Analysis

Image Quality

Using Rekognition Facial Analysis

Photo printing service can recommend the best

photos to their users

Online dating applications can improve their

match recommendations using face attributes

Retail businesses can understand the

demographics and sentiment of in-store

customers

Ad-tech services can display dynamic and

personalized content to customers

Facial Analysis - Use Case (Retail – In-store and Online)

Demographic and Sentiment Analysis

Female

Happy

Smiling

Male

No Facial Hair

Happy

Female

Sad

No Eyeglasses

Facial Analysis - Use Case (Retail – In-store and Online)

Demographic and Sentiment Analysis

AMAZON REDSHIFTAMAZON QUICKSIGHT AMAZON S3

Look Your Best All Day

Time for A New Look?

Pers

on A

Pers

on B

Sees

Sees

Facial Analysis - Use Case (Targeted Marketing)

Demographic and Sentiment Analysis

Facial Analysis - Use Case (Targeted Marketing)

Demographic and Sentiment Analysis

demographic and

sentiment attributes

Look Your Best All Day

Application

AMAZON

REDSHIFT

AMAZON

DYNAMODBAMAZON S3

log

demographic

profile updates

face image is collected

and analyzed AMAZON

REKOGNITION

DetectFaces

CompareFaces

Amazon Rekognition APIFace Comparison

Face-based user verification

Amazon Rekognition API

CompareFaces

{

"FaceMatches": [

{"Face": {"BoundingBox": {

"Height": 0.2683333456516266,

"Left": 0.5099999904632568,

"Top": 0.1783333271741867,

"Width": 0.17888888716697693},

"Confidence": 99.99845123291016},

"Similarity": 96

},

{"Face": {"BoundingBox": {

"Height": 0.2383333295583725,

"Left": 0.6233333349227905,

"Top": 0.3016666769981384,

"Width": 0.15888889133930206},

"Confidence": 99.71249389648438},

"Similarity": 0

}

],

"SourceImageFace": {"BoundingBox": {

"Height": 0.23983436822891235,

"Left": 0.28333333134651184,

"Top": 0.351423978805542,

"Width": 0.1599999964237213},

"Confidence": 99.99344635009766}

}

Face Comparison

IoT and camera manufacturers can integrate face-

based verification directly into their products

Application of face comparison in locating person of

interest for Public Safety

Hotels & hospitality businesses can provide seamless

access for guests and VIPs

Online exams or polls can verify presence of registered

person by comparing against image captured by

webcam.

Using Rekognition Face Comparison

Face Comparison – Use CaseFace-based Verification

AMAZON S3

Amazon Rekognition APIFacial Recognition

Index and Search faces in a collection

Index

Search

Collection

IndexFaces

SearchFacesByImage

Amazon Rekognition API

f7a3a278-2a59-5102-a549-a12ab1a8cae8

&

v1

02e56305-1579-5b39-ba57-9afb0fd8782d

&

v2

Face ID & face metadataFace

4c55926e-69b3-5c80-8c9b-78ea01d30690

&

v3tr

an

sfo

rme

d

sto

red

{

f7a3a278-2a59-5102-a549-a12ab1a8cae8,

02e56305-1579-5b39-ba57-9afb0fd8782d,

4c55926e-69b3-5c80-8c9b-78ea01d30690

}

IndexFace Collection

Amazon Rekognition API

Face

{

f7a3a278-2a59-5102-a549-a12ab1a8cae8,

02e56305-1579-5b39-ba57-9afb0fd8782d,

4c55926e-69b3-5c80-8c9b-78ea01d30690

}

SearchFacebyImage Collection

High dimensional

space search

Face ID

Facial Recognition

Using Facial Recognition

Family photo sharing apps can use face recognition

to group all faces of the same person in a family

Entertainment and news organizations can index

decades of archived images to find celebrities

Secure campuses / workplaces can use face search

to ensure all personnel in their facilities are

authorized to be there

Public safety teams can leverage face collections to

automate tracking of persons of interest

Facial Recognition - Use CasesIndexing faces into a collection

AMAZON S3

APPLICATION

Image Indexer

AMAZON

REKOGNITION

IndexFaces

Person Details

Application TableFace Collection

AWS LAMBDACAMERA

Live Frames

Facial Recognition - Use CasesSearch for similar faces in a face collection

AMAZON S3

Amazon Rekognition

Under the Hood

Amazon Rekognition – Feed Forward Inference

Layer 1 Layer 2 Layer n

…

Labrador

Dog

Beach

Outdoors

Cla

ssifie

r

Probability

Fully

Connected

Layer

Amazon Rekognition - Training Process

1. Data sourcing

2. Data Annotation

3. Annotation Validation

… and do this in an iterative process

based on feedback from customer,

system and QA validation

Amazon Rekognition - Training Process

Training

SimpleAutomated Quality Fast

Data Pipeline

Amazon Rekognition - Training Process

GlobalTrained

&

Trusted

AvailableElastic

Human Workforce

Amazon Rekognition - Training Process

DEMO

Benefits

Fully Managed AWS Integration Low CostProven Scalability Secure

Amazon Rekognition – Availability and Pricing

1. Released General Availability

2. At launch available in 3 regions,

1. US East (N. Virginia)

2. US West (Oregon)

3. EU (Ireland)

3. Pricing

• Pay as you go

• Free Tier – 5000 images per month for first 12 months

• Tiered Pricing designed

Amazon Rekognition – Pricing Details

Image Analysis TiersPrice per 1000

images processed

First 1 million images processed* per month $1.00

Next 9 million images processed* per month $0.80

Next 90 million images processed* per month $0.60

Over 100 million images processed* per month $0.40

*Images processed: For APIs with image as input, it’s the number of images analyzed. For APIs with no image input 1 API call = 1 image processed.

Summary

1. Fully managed and easy-to-use image recognition service

2. Four primary capabilities1. Object and Scene Detection

2. Facial Analysis

3. Face Comparison

4. Face Recognition

3. Integrated with AWS and AI Services

• Amazon S3

• Lex and Polly

4. Scalable and low cost

Getting Started

http://aws.amazon.com/rekognition/

Questions

Thank you

David Pearson