T HE S CHOOL OF A R T IFICI A L IN T ELLIGENCE

Intel® Edge AI for IoT

Developers

NANODEGREE SYLLABUS
 Overview

This Nanodegree Program is built in partnership with

Edge AI applications are revolutionizing

the IoT industry by bringing fast, intelligent Program Information
behavior to the locations where it is needed. TIME
In this Nanodegree program, you will 3 months
learn how to develop and optimize Edge Study 10 hours/week
AI systems, using the Intel® Distribution
LEVEL
of OpenVINO™ Toolkit. A graduate of this
Specialist
program will be able to:
• Leverage the Intel® Distribution PREREQUISITES
of OpenVINO™ Toolkit to fast-track • Intermediate Python
• Deep Learning
development of high-performance
• Command Line
computer vision and deep learning • OpenCV
inference applications.
• Run pre-trained deep learning models for HARDWARE/SOF T WARE
REQUIRED
computer vision on-prem.
Computer running a 64-bit
• Identify key hardware specifications of operating system that has
various hardware types (CPU, VPU, FPGA, 6th or newer generation of
and Integrated GPU). Intel® processor running
either Win, Ubuntu or (copy
• Utilize Intel® DevCloud for the Edge from Intel® Distribution of
to test model performance on various OpenVINO™ Toolkit).
hardware types (CPU, VPU, FPGA, and
LEARN MORE ABOUT THIS
Integrated GPU).
NANODEGREE
This program consists of 3 courses and 3 Contact us at
projects. Each project you build will be an enterpriseNDs@udacity.com.
opportunity to demonstrate what you’ve
learned in the course, and will demonstrate
to potential employers that you have skills in these areas.This program is comprised of 4 courses
and 4 projects. Each project you build will be an opportunity to demonstrate what you’ve learned
in the lesson, and will demonstrate to potential employers that you have skills in these areas.

2 THE SCHOOL OF ARTIFICIAL INTELLIGENCE

Our Classroom Experience

REAL-WORLD PROJECTS
Learners build new skills through industry-relevant
projects and receive personalized feedback from our
network of 900+ project reviewers. Our simple user
interface makes it easy to submit projects as often as
needed and receive unlimited feedback.

KNOWLEDGE
Answers to most questions can be found with
Knowledge, our proprietary wiki. Learners can search
questions asked by others and discover in real-time
how to solve challenges.

WORKSPACES
Learners can check the output and quality of their
code by testing it on interactive workspaces that are
integrated into the classroom.

QUIZZES
Understanding concepts learned during lessons is
made simple with auto-graded quizzes. Learners can
easily go back and brush up on concepts at anytime
during the course.

CUSTOM STUDY PLANS

Create a custom study plan to suit your personal
needs and use this plan to keep track of your progress
toward your goal.

PROGRESS TRACKER
Personalized milestone reminders help learners stay
on track and focused as they work to complete their
Nanodegree program.

Learn More at WWW.UDACITY.COM/ENTERPRISE INTEL® EDGE AI FOR IOT DEVELOPERS 3

 Learn with the Best

Stewart Christie Michael V ir go

COMMUNITY MANA G ER, IOT S ENI O R CU RRI CU L U M MANA GE R
DEVELOPER PROGRAM A T INTEL AT U D ACI TY
Stewart has been with Intel for the After beginning his career in business,
past 20+ years. He is an experienced Michael utilized Udacity Nanodegree
Technical Marketer with extensive programs to build his technical skills,
hardware and software knowledge. He eventually becoming a Self-Driving Car
has comprehensive understanding of Engineer at Udacity before switching roles
designing embedded electronic hardware to work on curriculum development for
and embedded firmware for various a variety of AI and Autonomous Systems
applications. programs.

Soham Chatterjee Vaidheeswaran Archana

GRAD UATE STU D ENT G RAD U ATE S TU D ENT
AT NTU AT NU 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 is
at Saama Technologies. He is currently currently pursuing her research in Deep
a Masters by Research student at NTU, Learning and Smart Grids, under Professor
Singapore. His research is on Edge Dipti Srinivasan. Archana is an Intel
Computing, IoT and Neuromorphic Software Innovator and a former Deep
Hardware Learning Engineer at Saama Technologies.

4 THE SCHOOL OF ARTIFICIAL INTELLIGENCE

Nanodegree Program Overview

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.

Project Deploy a People Counter App at the Edge

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 number of people in the frame, 2) time those people spent in frame, and 3) the total
number of people counted are sent to a web server; inference will be done on the local machine.

You will need to develop a method to compare the performance of their models before and after use of
the OpenVINO toolkit for optimization for edge deployment. You will also examine potential use cases for
their deployed people counter app.

LESSON TITLE LEARNING OUTCOMES

LEVERAGING PRE-
• Leverage a pre-trained model for computer vision inferencing.
TRAINED MODELS

THE MODEL • Convert pre-trained models into the framework agnostic intermediate
OPTIMIZER representation with the Model Optimizer.

THE INFERENCE • Perform efficient inference on deep learning models through the
ENGINE hardware-agnostic Inference Engine.

DEPLOYING AN • Deploy an app on the edge, including sending information through

EDGE APP MQTT, and analyze model performance and use cases.

Learn More at WWW.UDACITY.COM/ENTERPRISE INTEL® EDGE AI FOR IOT DEVELOPERS 5

 Nanodegree Program Overview

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 DevCloud 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.

Project Design a Smart Queuing System

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 requirements. You’ll use your
knowledge of hardware specifications to identify which hardware types might work, and then you’ll test
the application using the Intel® DevCloud 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.

LESSON TITLE LEARNING OUTCOMES

INTRODUCTION TO
• Describe the importance of selecting the right hardware and the
HARDWARE AT THE
process involved in doing so.
EDGE

• Identify the key specifications of Intel® CPUs and Integrated GPUs.

CPU AND
INTEGRATED GPU • Use the Intel® Devcloud for the Edge for running deep learning models
on the CPU and Integrated GPU.

6 THE SCHOOL OF ARTIFICIAL INTELLIGENCE

Nanodegree Program Overview

LESSON TITLE LEARNING OUTCOMES

• Identify the key specifications of Intel® VPUs.

VISION PROCESSING • Use the Intel® DevCloud for the Edge for running deep learning models
UNITS on the VPU.

• Use the MULTI Plugin to get more consistent performance.

• Identify the key specifications of Intel® FPGAs.

FIELD
• Use the Intel® DevCloud for the Edge for running deep learning models
PROGRAMMABLE
on the FPGA.
GATE ARRAYS
• Use the HETERO Plugin to enable efficient hardware utilization.

Learn More at WWW.UDACITY.COM/ENTERPRISE INTEL® EDGE AI FOR IOT DEVELOPERS 7

 Nanodegree Program Overview

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.

Project Build a Computer Pointer Controller

In this project, you will use models available in the 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 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 the extracted face. Using the head pose and facial
landmarks, you 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.

8 THE SCHOOL OF ARTIFICIAL INTELLIGENCE

Nanodegree Program Overview

LESSON TITLE LEARNING OUTCOMES

• Describe why Software Optimization is important.

INTRODUCTION • Identify the different fundamental optimization techniques.
TO SOFTWARE
OPTIMIZATION • Use different metrics to measure your model performance.

• Identify when and how to use optimization techniques.

• Calculate the number of operations in a model.

• Implement optimization techniques that improve performance by

REDUCING MODEL
reducing the number of model operations.
OPERATIONS
• •Use OpenVINO to measure the effects of different optimization
techniques on the performance of your model.

• Implement optimization techniques that improve performance by

REDUCING MODEL reducing the number of model operations.
SIZE • Use DL Workbench to quantize and measure the performance of your
model.

OTHER SOFTWARE • Use VTune Amplifier to measure hotspots in your application code.
OPTIMIZATION
TECHNIQUES • Package your application code and data.

Learn More at WWW.UDACITY.COM/ENTERPRISE INTEL® EDGE AI FOR IOT DEVELOPERS 9

 Our Nanodegree Programs Include:

Pre-Assessments Dashboard & Progress Reports

Our in-depth workforce assessments Our interactive dashboard (enterprise
identify your team’s current level of management console) allows administrators
knowledge in key areas. Results are used to to manage employee onboarding, track
generate custom learning paths designed course progress, perform bulk enrollments
to equip your workforce with the most and more.
applicable skill sets.

Industry Validation & Reviews Real World Hands-on Projects

Learners’ progress and subject knowledge Through a series of rigorous, real-world
is tested and validated by industry experts projects, your employees learn and
and leaders from our advisory board. These apply new techniques, analyze results,
in-depth reviews ensure your teams have and produce actionable insights. Project
achieved competency. portfolios demonstrate learners’ growing
proficiency and subject mastery.

10 THE SCHOOL OF ARTIFICIAL INTELLIGENCE

Our Review Process

Real-life Reviewers for Real-life Projects Vaibhav

Real-world projects are at the core of our Nanodegree programs UDACITY LEARNER
because hands-on learning is the best way to master a new skill.
Receiving relevant feedback from an industry expert is a critical part
of that learning process, and infinitely more useful than that from “I never felt overwhelmed while pursuing the
Nanodegree program due to the valuable support
peers or automated grading systems. Udacity has a network of over
of the reviewers, and now I am more confident in
900 experienced project reviewers who provide personalized and
converting my ideas to reality.”
timely feedback to help all learners succeed.
now at
CODING VISIONS INFOTECH
All Learners Benefit From:

Line-by-line feedback Industry tips and Advice on additional Unlimited submissions

for coding projects best practices resources to research and feedback loops

• Go through the lessons and work on the projects that follow

How it Works • Get help from your technical mentor, if needed
Real-world projects are • Submit your project work
integrated within the
• Receive personalized feedback from the reviewer
classroom experience,
making for a seamless • If the submission is not satisfactory, resubmit your project
review process flow. • Continue submitting and receiving feedback from the reviewer
until you successfully complete your project

About our Project Reviewers

Our expert project reviewers are evaluated against the highest standards and graded based on learners’ progress.
Here’s how they measure up to ensure your success.

900+ 1.8M 3 4.85

/5
Expert Project Projects Reviewed Hours Average Average Reviewer
Reviewers Our reviewers have Turnaround Rating
Are hand-picked extensive experience You can resubmit your Our learners love the
to provide detailed in guiding learners project on the same quality of the feedback
feedback on your through their course day for additional they receive from our
project submissions. projects. feedback. experienced reviewers.

Learn More at WWW.UDACITY.COM/ENTERPRISE INTEL® EDGE AI FOR IOT DEVELOPERS 11

Udacity © 2020
2440 W El Camino Real, #101
Mountain View, CA 94040, USA - HQ

For more information visit: www.udacity.com/enterprise

Udacity Enterprise Syllabus Intel Edge IoT Developer 19Aug2020 ENT