Jetson Nano Starter to Pro - A Computer Vision Course

Packt via Coursera

Go to class Write review

Details

Go to class

Provider

Coursera
Pricing

Paid Course
Languages

English
Certificate

Paid Certificate Available
Duration & workload

10 hours 21 minutes
Sessions

On-Demand
Subtitles

English

Found in

Overview

Coursera Flash Sale

40% Off Coursera Plus for 3 Months!

Grab it

This course features Coursera Coach! A smarter way to learn with interactive, real-time conversations that help you test your knowledge, challenge assumptions, and deepen your understanding as you progress through the course. In this course, you will explore the fascinating world of computer vision and artificial intelligence through the NVIDIA Jetson platform. By working through various modules, you'll gain a strong understanding of how to set up and optimize Jetson for AI tasks, with hands-on projects and practical applications of Jetson's capabilities. The course covers everything from basic image processing with OpenCV to advanced AI tools like YOLO, TensorRT, and DeepStream, helping you to develop cutting-edge computer vision applications. You will start by learning the basics of Jetson setup, moving on to installing essential libraries such as OpenCV and PyTorch, and applying them to create powerful image processing workflows. The course then takes you through object detection, deep learning, and AI optimization using tools like TensorRT, showcasing real-world applications like vehicle tracking and automatic number plate recognition. The final modules focus on integrating multiple cameras with DeepStream, enabling the creation of sophisticated surveillance systems. This course is ideal for those who wish to dive into the world of AI on edge devices, whether for robotics, surveillance, or other real-time applications. It is suitable for learners with basic programming knowledge and an interest in computer vision and AI development. By the end of the course, you will be equipped to build and deploy AI-powered computer vision systems using Jetson Nano.

Syllabus

Introduction to Jetson and Course Overview

In this module, we will introduce the NVIDIA Jetson platform and explain how to set up your Jetson device. We’ll also provide a detailed overview of the course, highlighting key topics and hands-on projects that will enhance your understanding of AI and robotics.

Comparison of Jetson and Its Variants Along with RPi+SD Card Flashing

In this module, we will compare Jetson with Raspberry Pi, helping you understand why Jetson is superior for AI applications. We will also guide you through flashing an SD card and selecting the optimal card for your device, ensuring a smooth setup.

Installing Libraries and Setting Up AI Computer - Explain Dependencies and Their Use

In this module, we will cover the installation of essential libraries such as OpenCV and PyTorch on Jetson. We’ll explore their functionalities and guide you through setting up an AI-ready environment for efficient project development.

Computer Vision OpenCV Basics on Jetson + Pytorch

In this module, we will dive into computer vision basics, demonstrating image manipulation with OpenCV and PyTorch. You will learn core techniques such as edge detection, image filtering, and geometric transformations to work with images effectively.

What is Object Detection + Yolo Object Detection

In this module, we will introduce you to the world of object detection and explore the YOLO algorithm, including its variants and their suitability for various use cases.

YOLO Object Detection on Custom Dataset (Number Plate Dataset)

In this module, we will walk you through the process of annotating a custom dataset, training a YOLO model, and applying it for number plate recognition. You will gain hands-on experience with both training and inference processes.

What is TensorRT? Setting Up Jetson for TensorRT

In this module, we will explore TensorRT, a critical tool for accelerating deep learning models on Jetson devices. We will also guide you through setting up your Jetson for TensorRT.

Optimizing YOLOX Model for Object Detection Using TensorRT

In this module, we will focus on optimizing the YOLOX object detection model using TensorRT. We will walk you through the conversion process, test the optimized model, and analyze performance improvements.

What is DeepStream and Theory?

In this module, we will introduce DeepStream, explaining how it works and how to set up the DeepStream SDK on your Jetson device. You will also learn how to utilize DeepStream for real-time AI applications.

Running DeepStream SDK and Setting up Multiple Cameras

In this module, we will demonstrate how to integrate multiple camera feeds into the DeepStream SDK. You will learn about video streaming protocols and how to perform object detection across multiple cameras in real-time.

App 1 Car detection + Tracking + Counting

In this module, we will guide you through the process of implementing a vehicle detection and tracking system. You will learn how to configure the application and evaluate its performance in real-world scenarios.

Automatic Number Plate Recognition with Paddle OCR

In this module, we will cover the process of training a custom object detection model for number plate recognition. You will use Roboflow, Google Colab, and Paddle OCR to build and deploy an effective ANPR system.

Pose Estimation Method 1: PoseNet

In this module, we will introduce pose estimation and demonstrate how to use PoseNet on Jetson for accurate pose detection. We will also explore techniques to enhance PoseNet with Darknet and Mediapipe.

Pose Estimation Method 2: PoseNet

In this module, we will dive deeper into PoseNet for pose estimation. You will learn the full implementation process, gaining a thorough understanding of how to apply PoseNet effectively.

DeepFake face classification

In this module, we will introduce you to DeepFake technology, discussing its ethical concerns and providing techniques for detecting DeepFake content using AI-based classification methods.

Face Recognition and Attendance: Clock in, clock out.

In this module, we will explain the role of face recognition in attendance management systems. You will learn how to implement and fine-tune such a system for accurate tracking of attendance.