YOLOv4 Object Detection Course
What you'll learn
- The basics about YOLOv4
- Installing all the pre-requisites including Python, OpenCV, CUDA and Darknet
- You will be able to detect objects on images
- Implement YOLOv4 Object detection on videos
- Creating your own social distancing monitoring app
- Basic python programming skills
- Mid to high range PC or laptop with Windows 10 operating system
- Enthusiasm to learn AI
- CUDA enabled GPU (Graphics Card)
- Basic Understanding of Computer Vision
I started out wanting to learn AI Object Detection in Computer Vision...
... I used to check a lot of GitHub repos, they were very vague and required for me to be competent in software development/programming and understand all of the jargon –
Now even though I have a masters degree in electronic engineering (M.Eng). It was still challenging for me to figure out. I had a lot of questions like...
...What to do to get my code working?
Do I have the right hardware
Windows or Linux – If linux, do I use Ubuntu, Red Hat, CentOS, ROS
If Ubuntu, what version 16.04, 18.04, What kernel do I need?
If I am training, what format does my dataset need to be in?
Do I use Python or C++
If python What dependencies do I need?
Which frameworks do I use? PyTorch, TensorFlow 1.0 or 2.0
What commands do I type to infer or train a convolutional neural network
How big my dataset needs to be?
How do I run on GPU, and does my GPU support the framework?
How to train YOLOv4
How create cross platform apps using Yolov4 and PyQt
I was unsure of what to do. Sometimes I would look at the instructions and because the instructions were so vague, I would skip to the next repo and the next, until I found one that resonates with me or one that had a clear set of instructions that I could understand and follow, or had a video tutorial on it. And video tutorials on this particular topic are very scarce.
The other problem was, I would follow the instructions, but I would run in trivial issues, like not having the correct dependencies or I did not have the correct hardware or OS etc. When things don’t work. This would beat me down and make me loose confidence of whether or not this repository would work. Now I had 2 options, I could either spend tons of hours searching the web to debug the issue or move on to the next repo which also may or may not work.
Then, I thought, if me with a masters degree in electronic engineering had all these issues with getting started in AI, surely other people would be having this same issue as me. People such as:
non-programmers/non computer science ,
Hobbyists, Students, researcher, employees.
People starting out in AI....
The YOLOv4 Object Detection Course
When YOLOv4 was released in April 2020, my team and I worked effortlessly to create a course in which will help you implement YOLOv4 with ease. We created this Nano course in which you will learn the basics and get started with YOLOv4. This is all about getting object detection working with YOLOv4 in your windows 10 PC.
You will learn how to install all the dependencies, including Python, CUDA and OpenCV. Once you’ve managed to compile it successfully, we go on to execute YOLOv4 on images and videos. Then to ensure that you understand whats going on, we delve deeper into the darknet python script and show you how to also run YOLOv4 on a webcam.
Within this nano-course, we shall also create our first weapon against COVID-19 which is our social distancing monitoring app. Which essentially monitors the physical distance between people to ensure that they’re keeping safe distancing from each other. It also displays the number of people at risk at any given time
The YOLOv4 Course provides you with a gentle introduction to the world of computer vision with YOLOv4, first by learning how to install darknet, building libraries for YOLOv4 all the way to implementing YOLOv4 on images and videos in real-time.
From here you will even solve current and relevant real-world problems by building your own social-distancing monitoring app.
Please ensure that you have the following:
Basic understanding of Computer Vision
Python Programming Skills
Mid to high range PC/ Laptop
CUDA-enabled GPU - Important*
Imagine, if a week from now, once you have completed this course, that you are able to implement and implement your own Convolutional Neural Networks (CNN's) with YOLOv4 object detection pre-trained model. Imagine all the applications you could do with these skills!
You could be take your new found expertise and be:
Solving real world problems,
Freelancing AI projects,
Getting that job/opportunity in AI,
Tackling your research guns blazing!
Saving time, money, &
Wishing you had done this course sooner.
The world is your oyster... Ask yourself...What cool things would you do once you have skills in AI?
So what are you waiting for?
Who this course is for:
- Are a computer vision developer that utilizes AI and are eager to level-up your skills.
- Have experience with machine learning and want to break into neural networks or AI for visual understanding.
- Are a scientist looking to apply deep learning + computer vision algorithms to your research.
- Are a university student and want more than your university offers (or want to get ahead of your class).
- Utilize computer vision algorithms in your own projects but have yet to try deep learning.
- Used AI in projects before, but never in the context of analysis of visual perception.
- Write Python/ML code at your day job and are motivated to stand out from your coworkers.
- Are a "AI hobbyist" who knows how to program and wants to tinker with DIY projects using computer vision.
- You understand that this requires hard work and patience to get the right skills. You understand that you’re going to get any results overnight.
- You’re someone that believes in taking action. You watch the material and then you actually APPLY it.
So a bit about me, Ritesh Kanjee: I've graduated from University of Johannesburg as an Electronic Engineer with a Masters in Image Processing and 8 years ago I started my online school called Augmented Startups where I have over 97'000 subscribers on YouTube and over 60'000 students on Augmented AI Bootcamp/Udemy.
I’ve worked with popular tools such as TensorFlow Keras, Open CV, and PyTorch and I’ve also produced High ranking tutorials that feature on Google and YouTube. My Machine Learning Series is also one of the most viewed videos, over 300 thousand views and you’ll find them ranked right at the top on YouTube search results.
From my tutorials, I have received a lot of great feedback and testimonials from students all around the world, I will share those reviews towards the end of the video
And I have also presented at international conferences and meetups in AI. For industry standard AI, I have partnered up with Geeky Bee AI who are Experts in the field in AI and Deep Learning and have experience developing AI apps for real world applications.
Geeky Bee AI Pvt Ltd (The Artificial Intelligence Solution Provider) offers development in the field of computer vision, deep learning and automation to solve complex challenges for clients across the world.
Geeky Bee was founded in 2018 by a group of young experienced professionals with desire to build affordable products with high-end technologies and bring it to every individual’s doorstep. Its headquarters is located in Ahmedabad, India.
Languages: C, C++, Python, Java
Tools: Visual Studio, Pycharm, Anaconda, Jupyter Notebook, Android Studio, Arduino IDE
Cloud: AWS, Onepanel, Azure, Google Cloud, Floydhub
OS: Windows, Android, Ubuntu, Mac High Sierra, ROS, Cent OS, Kali Linux
Libraries & Framework: OpenCV, DLib, TensorFlow, MACE, Tesseract, NLTK, Spacy, Keras, Caffee, Pytorch, PipeCNN, .Flask
Models: YOLO, SSD, CNN, RCNN, MobileNet, Alexnet, ImageNet, Inception V3, DeepLab
Hardware: Raspberry Pi, Node MCU, Xilinx FPGAs