The Practical Evolution of the Auto-Tagging Technology as a Service - - PowerPoint PPT Presentation

▶
the practical evolution of the auto tagging technology as
SMART_READER_LITE
LIVE PREVIEW

The Practical Evolution of the Auto-Tagging Technology as a Service - - PowerPoint PPT Presentation

Aug 04, 2023 531 likes •955 views

The Practical Evolution of the Auto-Tagging Technology as a Service Georgi Kadrev @georgikadrev Imagga Technologies @imagga GTC 17, May10th In numbers more than more than in more than 200 8.9K 82 Businesses Developers Countries

slide-1
SLIDE 1

The Practical Evolution

  • f the Auto-Tagging

Technology as a Service

Georgi Kadrev @georgikadrev Imagga Technologies @imagga GTC ’17, May10th

slide-2
SLIDE 2

more than

200

Businesses more than

8.9K

Developers in more than

82

Countries Worldwide

In numbers

slide-3
SLIDE 3

3B

New photos shared

Every Day

more than

slide-4
SLIDE 4

All the buzz words…

Artificial Intelligence Machine Learning Deep Neural Network Convolutional Neural Network Deep Learning

slide-5
SLIDE 5

Why not

  • rganize photos

using these via an auto-tagging API?!

slide-6
SLIDE 6

Image Recognition as Service

Submit an image Get a list of tags or categories Do whatever you need with them

3 2 1

slide-7
SLIDE 7

Frameworks Services vs.

slide-8
SLIDE 8
  • CudaConvnet
  • Caffe
  • Torch
  • Theano
  • MS Cognitive Toolkit
  • TensorFlow
  • Caffe2

Popular Frameworks

slide-9
SLIDE 9
  • Collect data
  • Train and optimize
  • Deploy and handle scale
  • Support and maintain
  • Keep innovating

Frameworks

DIY (Do It Yourself)

slide-10
SLIDE 10
  • Zero time to market
  • No backend operations needed
  • Pay as you go pricing

Services

Utilize a service

slide-11
SLIDE 11

macaw beak parrot bird tropical wildlife blue black light blue navy blue

  • range

10%

13%

100%

10%

100% 100%

Keyword and Color Tagging

slide-12
SLIDE 12

12

Beaches & Seaside Sunrises & Sunsets Nature & Landscape Events & Parties

Category Tagging

slide-13
SLIDE 13

NSFW

(Not Safe For Work)

  • Safe
  • Underwear
  • Not safe

Content Moderation Tagging

slide-14
SLIDE 14

Underlying Technology

Neural Networks

deep learning + feature extraction
 + semantic expansion

Automated Analysis

  • f raster/pixel data

Image Tagging

3,000+ objects 20,000+ terms/concepts

slide-15
SLIDE 15

Traditional Image Processing

and/or crowd-sourced

Deep Learning Based

&

slide-16
SLIDE 16
  • Amazon MT
  • TagCow
  • Imagga
  • Imense
  • CamFind (Cloudsight)

Traditional Image Processing

slide-17
SLIDE 17
  • Imagga
  • Clarifai
  • Rekognition (Orbeus)
  • Cloudsight
  • Metamind
  • MS Cognitive Services
  • IBM Watson
  • Google Cloud Vision
  • Amazon Rekognition (acquired Orbeus)
  • Salesforce Einstein (acquired Metamind)

Deep Learning Based

slide-18
SLIDE 18
  • Detect faces
  • Extract text
  • Extract colors
  • Analyze composition
  • Classify/categorize scenes
  • Recognize the main object
  • Suggest multiple keywords
  • Generate textual descriptions
  • Recognize quality and art value
  • Localize multiple objects

Practical Features

slide-19
SLIDE 19

Detect Faces

Image source: Wikipedia, CC3 license

slide-20
SLIDE 20

Extract Text

Image source: http://simon-tanner.blogspot.com/2015/06/text-capture-and-optical-character.html

slide-21
SLIDE 21

Extract Colors

blue black light blue navy blue

  • range
slide-22
SLIDE 22

Analyze Composition

slide-23
SLIDE 23

23

Beaches & Seaside Sunrises & Sunsets Nature & Landscape Events & Parties

Classify/Categorize Scenes

slide-24
SLIDE 24

Recognize The Main Object

parrot

100%

slide-25
SLIDE 25

Suggest Multiple Keywords

macaw beak parrot bird tropical wildlife

10%

13%

100%

10%

100% 100%

slide-26
SLIDE 26

Localize Multiple Objects

Image source: https://www.linkedin.com/pulse/object-detection-using-deep-learning-advanced-users-part-1-sinhal

slide-27
SLIDE 27

Generate Textual Descriptions

Image captured from: http://cs.stanford.edu/people/karpathy/deepimagesent/

slide-28
SLIDE 28

Recognize Quality and Art Value

Image source: EyeEm Vision demo page

slide-29
SLIDE 29
  • Internal organization of photos
  • Organization of photos for sale
  • Organization of personal photos
  • Content moderation
  • Marketing and advertising
  • Analytics and profiling
  • Content recommendation

Practical Use-cases

slide-30
SLIDE 30
  • Cloud Services / Telcos
  • Social Media Monitoring
  • Contextual Advertising
  • Digital Asset Management
  • Image Processing Platforms
  • Smart Devices and Installations

Use-cases

slide-31
SLIDE 31
  • Is 95% of cloud storage photos and videos? YES!
  • Does Google, Apple and Amazon have image recognition? YES!
  • Do you have it? NO!
  • Do you want to leave them ahead or you want to have it right now? NOW!
  • What do you get? Increased cloud retention and lock in!
  • Does it really work? YES!
  • How do we know? Swisscom and 200+ more happy paying customers!

Use case: Cloud Service Providers/Telcos

slide-32
SLIDE 32

Use-case: Swisscom

Enhancing Swisscom myCloud with automated image organization.

Tagging API

Technologies used:

Categorization API

slide-33
SLIDE 33

Use-case: Unsplash

Unsplash

Providing powerful image search capabilities.

Reduces/replace manual tagging and enhances search.

Tagging API

Technologies used:

slide-34
SLIDE 34

Use-case: Tavisca

Building a custom hotel classifier.

Tagging API

Technologies used:

Custom Training

Automates classification and improving browsing experience

slide-35
SLIDE 35

Use-case: Tavisca

KIA K5 (Optima) Creative Campaign

Very precise personalized targeting.

Tagging API

Technologies used:

Color Extraction API

slide-36
SLIDE 36

Challenges

  • Definition of the scope
  • Data
  • Even more data
  • True learning from data
  • Learning from private data
slide-37
SLIDE 37

Upcoming Imagga features

  • Positional object detection
  • Official face recognition support
  • Out-of-the-box on-premise packaging
  • Official video support
  • Logo and landmark recognition
  • Better multi-language support
slide-38
SLIDE 38

On-premise solution available

  • World first
  • Ultimate Privacy
  • Enterprise class
  • Easy deployment
  • Custom categorization option
  • Prepaid or pay-as-you go volume-based license
  • Standard and advanced support per machine

On-premise Solution LAUNCHED TODAY!

slide-39
SLIDE 39

Integration & Business Model

On-premise Solution Cloud API Platform

Monthly Subscription Self-service Volume License
 S&M

slide-40
SLIDE 40

Giants:

  • Google Cloud Vision
  • AWS Rekognition
  • Microsoft Cognitive API
  • IBM Watson Vision
  • Salesforce Einstein

Imagga:

  • On-premise option
  • Professional custom training

Competitive Landscape Notable Startups:

  • Clarifai
  • EyeEm (Vision)
  • CloudSight
slide-41
SLIDE 41

Thank You!

api@imagga.com twitter.com/imagga facebook.com/imagga