NANODEGREE PROGR AM SYLL ABUS

Intel® Edge AI for IoT

Developers

Need Help? Speak with an Advisor: www.udacity.com/advisor

Overview
Edge AI applications are revolutionizing the IoT industry by bringing fast, intelligent behavior to the locations
where it is needed. In this Nanodegree program, you will learn how to develop and optimize Edge AI
systems, using the Intel® Distribution of OpenVINO™ Toolkit. A graduate of this program will be able to:

• Leverage the Intel® Distribution of OpenVINO™ Toolkit to fast-track development of high-performance

computer vision and deep learning inference applications.
• Run pre-trained deep learning models for computer vision on-prem.
• Identify key hardware specifications of various hardware types (CPU, VPU, FPGA, and Integrated GPU).
• Utilize Intel® DevCloud for the Edge to test model performance on various hardware types (CPU, VPU,
FPGA, and Integrated GPU).

This program consists of 3 courses and 3 projects. Each project you build will be an opportunity to
demonstrate what you’ve learned in the course, and will demonstrate to potential employers that you have
skills in these areas.

I N CO L L A B O R AT I O N W I T H

Estimated Time: Prerequisites:

3 Months at Intermediate
10 hours / week Python, Deep
Learning,
Command Line,
OpenCV

Flexible Learning: Need Help?

Self-paced, so udacity.com/advisor
you can learn on Discuss this program
the schedule that with an enrollment
works best for you. advisor.

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 2
Course 1: Edge AI Fundamentals with OpenVINO™
Leverage a pre-trained model for computer vision inferencing. You will convert pre-trained models into the
framework agnostic intermediate representation with the Model Optimizer, and perform efficient inference on
deep learning models through the hardware-agnostic Inference Engine. Finally, you will deploy an app on the edge,
including sending information through MQTT, and analyze model performance and use cases

In this project, you will utilize the Intel® Distribution of the
OpenVINO™ Toolkit to build a People Counter app for inference at
the edge. You will investigate different pre-trained models for person
detection, and then convert the best model for optimized inference.
The model will be deployed on the edge, such that only data on 1) the
Course Project
number of people in the frame, 2) time those people spent in frame,
Deploy a People and 3) the total number of people counted are sent to a web server;
Counter App at the Edge inference will be done on the local machine.

You will need to develop a method to compare the performance

of your models before and after use of the Intel® Distribution of
OpenVINO™ Toolkit for optimization for edge deployment. You will also
examine potential use cases for your deployed people counter app.

LEARNING OUTCOMES

Leveraging Pre- • Leverage a pre-trained model for computer vision

LESSON ONE
Trained Models inferencing

• Convert pre-trained models into the framework agnostic

LESSON TWO The Model Optimizer
intermediate representation with the Model Optimizer

• Perform efficient inference on deep learning models

LESSON THREE The Inference Engine
through the hardware-agnostic Inference Engine

• Deploy an app on the edge, including sending information

Deploying An Edge
LESSON FOUR through MQTT, and analyze model performance and use
App
cases.

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 3
Course 2: Hardware for Computer Vision
Deep Learning Application Deployment
Grow your expertise in choosing the right hardware. Identify key hardware specifications of various
hardware types (CPU, VPU, FPGA, and Integrated GPU). Utilize the Intel® DevCloud for the Edge to test
model performance and deploy power-efficient deep neural network inference on on the various hardware
types. Finally, you will distribute workload on available compute devices in order to improve model
performance.

In this project, you will be given a real-world scenario of building a

queuing system for three different clients in three different industry
sectors. The sectors will consist of retail, manufacturing, and
transportation. Each client will have their own set of constraints and
Course Project requirements. You’ll use your knowledge of hardware specifications
Design a Smart Queuing to identify which hardware types might work, and then you’ll test
System the application using the Intel® DevCloud for the Edge to see which
hardware performs best. Finally, after reviewing your test results
and considering the constraints and requirements of the client, you
will propose a hardware solution and justify your selection.

LEARNING OUTCOMES

Introduction to
• Describe the importance of selecting the right hardware and
LESSON ONE Hardware at the
the process involved in doing so
Edge

• Identify the key specifications of Intel® CPUs and Integrated

CPU and Integrated GPUs
LESSON TWO
GPU • Use the Intel® Devcloud for the Edge for running deep learning
models on the CPU and Integrated GPU

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 4
 • Identify the key specifications of Intel® VPUs
Vision Processing • Use the Intel® DevCloud for the Edge for running deep learning
LESSON THREE
Units models on the VPU
• Use the MULTI Plugin to get more consistent performance

• Identify the key specifications of Intel® FPGAs

Field Programmable • Use the Intel® DevCloud for the Edge for running deep learning
LESSON FOUR
Gate Arrays models on the FPGA
• Use the HETERO Plugin to enable efficient hardware utilization

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 5
Course 3: Optimization Techniques and Tools
for Computer Vision Deep Learning Applications
Learn how to optimize your model and application code to reduce inference time when running your
model at the edge. Use different software optimization techniques to improve the inference time of your
model. Calculate how computationally expensive your model is. Use DL Workbench to optimize your
model and benchmark the performance of your model. Use a VTune amplifier to find and fix hotspots in
your application code. Finally, package your application code and data so that it can be easily deployed to
multiple devices.

In this project, you will use models available in the Intel®

Distribution of OpenVINO™ Toolkit to control your computer
pointer using your eye gaze. You will first have to identify faces and
extract a face from an input video stream captured from a webcam
Course Project or a video file. Then you will need to extract facial landmarks and
also use a head pose estimation model to find the orientation of
Build a Computer
the extracted face. Using the head pose and facial landmarks, you
Pointer Controller will find the orientation of the eye gaze using a gaze estimation
model. Finally, you will need to move the mouse pointer in the
direction of the eye gaze. This project will demonstrate your ability
to run multiple models in the same machine, and coordinate and
optimize the flow of data between those models.

LEARNING OUTCOMES

• Describe why Software Optimization is important

Introduction
• Identify the different fundamental optimization techniques
LESSON ONE to Software
• Use different metrics to measure your model performance
Optimization
• Identify when and how to use optimization techniques

• Calculate the number of operations in a model

• Implement optimization techniques that improve
Reducing Model performance by reducing the number of model operations
LESSON TWO
Operations • Use the Intel® Distribution of OpenVINO™ Toolkit to measure
the effects of different optimization techniques on the
performance of your model

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 6
 LEARNING OUTCOMES

• Implement optimization techniques that improve

Reducing Model performance by reducing the number of model operations
LESSON THREE
Size • Use DL Workbench to quantize and measure the performance
of your model

Other Software • Use VTune Amplifier to measure hotspots in your application

LESSON FOUR Optimization code
Techniques • Package your application code and data

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 7
Our Classroom Experience
REAL-WORLD PROJECTS
Build your skills through industry-relevant projects. Get
personalized feedback from our network of 900+ project
reviewers. Our simple interface makes it easy to submit
your projects as often as you need and receive unlimited
feedback on your work.

KNOWLEDGE
Find answers to your questions with Knowledge, our
proprietary wiki. Search questions asked by other students
and discover in real-time how to solve the challenges that
you encounter.

STUDENT HUB
Leverage the power of community through a simple, yet
powerful chat interface built within the classroom. Use
Student Hub to connect with your technical mentor and
fellow students in your Nanodegree program.

WORKSPACES
See your code in action. Check the output and quality of
your code by running them on workspaces that are a part
of our classroom.

QUIZZES
Check your understanding of concepts learned in the
program by answering simple and auto-graded quizzes.
Easily go back to the lessons to brush up on concepts
anytime you get an answer wrong.

CUSTOM STUDY PLANS

Work with a mentor to create a custom study plan to suit
your personal needs. Use this plan to keep track of your
progress toward your goal.

PROGRESS TRACKER
Stay on track to complete your Nanodegree program with
useful milestone reminders.

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 8
Learn with the Best

Stewart Christie Michael Virgo

CO M M U N I T Y M A N A G E R , I OT S E N I O R C U R R I C U LU M M A N AG E R
D E V E LO P E R P R O G R A M AT I N T E L ® AT U DAC I T Y

Stewart is a Technical Evangelist for Intel®, After beginning his career in business,
responsible for running workshops, Michael utilized Udacity Nanodegree
creating content, and supporting the programs to build his technical skills,
developer community in IoT. He is skilled eventually becoming a Self-Driving Car
in developing applications that interface Engineer at Udacity before switching roles
hardware with software for computer to work on curriculum development for
vision, robotics, and language processing. a variety of AI and Autonomous Systems
programs.

Soham Chatterjee Vaidheeswaran Archana

G R A D UAT E S T U D E N T G R A D UAT E S T U D E N T
AT N T U AT N U S

Soham is an Intel® Software Innovator Archana is a graduate student at the

and a former Deep Learning Researcher National University of Singapore. She
at Saama Technologies. He is currently is currently pursuing her research in
a Masters by Research student at NTU, Deep Learning and Smart Grids, under
Singapore. His research is on Edge Professor Dipti Srinivasan. Archana is
Computing, IoT and Neuromorphic an Intel® Software Innovator and a
Hardware. former Deep Learning Engineer at Saama
Technologies.

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 9
All Our Nanodegree Programs Include:

EXPERIENCED PROJECT REVIEWERS

RE VIE WER SERVICES

• Personalized feedback & line by line code reviews

• 1600+ Reviewers with a 4.85/5 average rating
• 3 hour average project review turnaround time
• Unlimited submissions and feedback loops
• Practical tips and industry best practices
• Additional suggested resources to improve

TECHNICAL MENTOR SUPPORT

M E N TO R S H I P S E R V I C E S

• Questions answered quickly by our team of

technical mentors
• 1000+ Mentors with a 4.7/5 average rating
• Support for all your technical questions

PERSONAL CAREER SERVICES

C A R E E R CO A C H I N G

• Personal assistance in your job search

• Monthly 1-on-1 calls
• Personalized feedback and career guidance
• Access to Udacity Talent Program used by our
network of employers to source candidates
• Advice on negotiating job offers
• Interview preparation
• Resume services
• Github portfolio review
• LinkedIn profile optimization

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 10
Frequently Asked Questions
PROGR AM OVERVIE W

WHY SHOULD I ENROLL?

70% of data being created is at the edge, and only half of that will go to the
public cloud; the rest will be stored and processed at the edge, which requires
a different kind of developer. Demand for professionals with the Edge AI
skills will be immense, as the Edge Artificial Intelligence (AI) software market
size is forecasted to grow from $355 Million in 2018, to $1,152 billion by 2023,
at an Annual Growth Rate of 27%. (MarketsandMarkets) In the Edge AI for
IoT Developers Nanodegree program, you’ll leverage the potential of edge
computing and use the Intel® Distribution of OpenVINO™ Toolkit to fast-track
development of high-performance computer vision and deep learning inference
applications.

Computer Vision is a fast-growing technology being deployed in nearly every

industry from factory floors to amusement parks to shopping malls, smart
buildings, and smart homes. It is also driving the evolution of machine learning
and human interactions with intelligent systems. Additional applications include
drones, security cameras, robots, facial recognition on cell phones, self-driving
vehicles, and more, which means these industries and more all need developers
with computer vision and deep learning IoT experience.

WHAT JOBS WILL THIS PROGRAM PREPARE ME FOR?

This Nanodegree program will prepare you for roles such as IoT Developer, IoT
Engineer, Deep Learning Engineer, Machine Learning Engineer, AI Specialist,
VPU/CPU/FPGA Developer and more for companies and organizations looking
to innovate their hardware on the Edge.

HOW DO I KNOW IF THIS PROGRAM IS RIGHT FOR ME?

If you are an enterprise developer and/or professional developer interested in
advanced learning, specifically deep learning and computer vision, this program
is right for you.

Additionally if you have a background as an IoT Application Prototyper, IoT

Application Implementer, IoT System Prototyper, or an IoT System Implementer,
or in heterogeneous architectures as an Device Developer, Application
Prototyper, Algorithm Developer, Solution Developer, or in security as an
Architect/Planner, Security Specialist, or a Protocol Implementer, this program is
a good fit.

WHAT IS EDGE AI? WHAT ARE SOME APPLICATIONS OF THIS TECHNOLOGY?

Edge Computing runs processes locally on the device itself, instead of running
them in the cloud. This reduced computing time allows data to be processed
much faster, removes the security risk of transferring the data to a cloud-based
server, and reduces the cost of data transfer, as well as the risks of bandwidth
outages disrupting performance.

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 11
FAQs Continued
Computer vision and AI at the edge are becoming instrumental in powering
everything from factory assembly lines and retail inventory management,
to hospital urgent care medical imaging equipment like X-ray and CAT scans.
Drones, security cameras, robots, facial recognition on cell phones, self-driving
vehicles, and more all utilize this technology as well.

According to IEEE Innovation at Work, “By 2020, approximately 20+ billion devices
will likely be connected via the Internet of Things (IoT), creating incredible
amounts of data every minute. The time it takes to move data to the cloud,
perform service on it and then move it back to devices is far too long to meet the
increasing needs of the IoT. Unlike cloud computing, which relies on a single data
center, edge computing works with a more distributed network, eliminating the
round-trip journey to the cloud and offering real-time responsiveness and local
authority. It keeps the heaviest traffic and processing closest to the end-user
application and devices – smartphones, tablets, home security systems, and
more – that generate and consume data. This dramatically reduces latency and
leads to real-time, automated decision-making.” (IEEE)

WHAT IS THE INTEL® DEVCLOUD FOR THE EDGE?

The Intel® DevCloud for the Edge allows you to actively prototype and
experiment with AI workloads for computer vision on Intel hardware.

You have full access to hardware platforms hosted in our cloud environment,
designed specifically for deep learning. You can test the performance of your
models using the Intel® Distribution of OpenVINO™ Toolkit and combinations of
CPUs, GPUs, VPUs such as the Intel® Neural Compute Stick 2 (NCS2) and FPGAs,
such as the Intel® Arria® 10. The Intel® DevCloud for the Edge contains a series
of Jupyter* notebook tutorials and examples preloaded with everything you
needed to quickly get started.

This includes trained models, sample data and executable code from the Intel®
Distribution of OpenVINO™ Toolkit as well as other tools for deep learning.
These notebooks are designed to help you quickly learn how to implement deep
learning applications to enable compelling, high-performance solutions. Intel®
has AI hardware waiting for your prototyping of edge inference jobs.

No hardware setup is required on your end. The Intel® DevCloud for the Edge
utilizes Jupyter* Notebooks to execute code directly within the Web browser.
Jupyter* is a browser-based development environment which allows you to run
code and immediately visualize results. You can prototype innovative computer
vision solutions in our cloud environment, then execute your code on any of
Intel’s® available combination of hardware resources.

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 12
FAQs Continued
WHAT MAKES THE INTEL® EDGE AI FOR IOT DEVELOPERS NANODEGREE
PROGRAM UNIQUE?
The Intel® Distribution of OpenVINO™ Toolkit is for developers looking to
deploy deep learning models on hardware with Intel chips. Students will
be able to interact with Intel’s IoT development platform to optimize the
performance of their hardware using the DL Workbench. Through Udacity’s
interactive workspaces, you’ll be able to send jobs to the Intel® DevCloud for
the Edge and see how different hardware performs in real time.

Deploying AI models on the Edge requires a particular set of tools that

providers such as Intel have built. Through Udacity’s hands-on exercises that
integrate with Intel’s platform, students will be able to actually practice testing
AI model performance on hardware without needing access to the hardware.

The Intel® DevCloud for the Edge is a cloud-based platform that lets you
deploy machine learning models on hardware in the cloud before you purchase
the actual hardware so you test and compare the performance of different
hardware.

ENROLLMENT AND ADMISSION

DO I NEED TO APPLY? WHAT ARE THE ADMISSION CRITERIA?

There is no application. This Nanodegree program accepts everyone,
regardless of experience and specific background.

WHAT ARE THE PREREQUISITES FOR ENROLLMENT?

To succeed in this program, students should have the following:
• Intermediate knowledge of programming in Python
• Experience with training and deploying deep learning models
• Familiarity with different DL layers and architectures (CNN based)
• Familiarity with the command line (bash terminal)
• Experience using OpenCV

IF I DO NOT MEET THE REQUIREMENTS TO ENROLL, WHAT SHOULD I DO?

There are a few courses that can help prepare you for the program:
• For Python Experience: AI Programming with Python
• For Deep Learning Experince: Deep Learning
• For AI Modeling: Intro to Machine Learning with Pytorch or
Intro to Machine Learning with TensorFlow
• For Computer Vision Experience: Computer Vision

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 13
FAQs Continued
TUITION AND TERM OF PROGR AM

HOW IS THIS NANODEGREE PROGRAM STRUCTURED?

The Intel® Edge AI for IoT Developers Nanodegree program is comprised
of content and curriculum to support three (3) projects. We estimate that
students can complete the program in three (3) months, working 10 hours per
week.

HOW LONG IS THIS NANODEGREE PROGRAM?

Access to this Nanodegree program runs for the length of time specified in
the payment card on the Nanodegree program overview page. If you do not
graduate within that time period, you will continue learning with month to
month payments. See the Terms of Use for other policies around the terms of
access to our Nanodegree programs.

I HAVE GRADUATED FROM THE C++ NANODEGREE PROGRAM, WHERE

SHOULD I GO FROM HERE?
Many of our graduates continue on to our Machine Learning Engineer
Nanodegree program, and after that, to the Self-Driving Car Engineer and
Artificial Intelligence Nanodegree programs.

CAN I SWITCH MY START DATE? CAN I GET A REFUND?

Please see the Udacity Nanodegree program FAQs for policies on enrollment in
our programs.

S O F T WA R E A N D H A R D WA R E

WHAT SOFTWARE AND VERSIONS WILL I NEED IN THIS PROGRAM?

You will need a computer running a 64-bit operating system that has 6th or
newer generation of Intel® processor running either Win, Ubuntu or (copy from
Intel® DevCloud for the Edge OpenVINO™).

You will also need to:

• Install Intel® Distribution of OpenVINO™ Toolkit (OpenVINO™ 2020.1) on
your local environment.
• Install Intel’s® Deep Learning Workbench (Please note that DL Workbench
does not currently support Windows 10 Home Edition. We recommend
students either upgrade to Windows 10 Professional or use a Linux based
system.)

Need Help? Speak with an Advisor: www.udacity.com/advisor Intel® Edge AI for IoT Developers | 14