Overview
OpenCV.org, in partnership with Big Vision LLC (owner of LearnOpenCV.com) has launched a Kickstarter campaign to create 3 Computer Vision courses. There are many benefits of buying these courses during the Kickstarter.
- 50% discounts on all courses and additional discounts if you buy multiple courses ( called Bundles ).
- 12 months for completing the projects and getting the certification (will be 6 months after the campaign).
- Access to both the C++ and Python courses without any additional cost.
- An option to buy all future courses at a fixed price ( Ultimate Bundle ).
Please use this link to pre-book your seat in the courses at a 50% DISCOUNT.
There is a lot of educational material available online for computer vision, machine learning, and AI. However, the material is not organized for beginners to learn effectively. Often the focus is on showing off a cool shiny application without covering the fundamentals.
Because of this reason, at OpenCV.org, we have adopted a mission to not only build the best open source computer vision library in the world, but also to create the most comprehensive online courses to educate a global workforce.
Kickstarter Campaign
OpenCV Foundation is a non-profit organization. We need help from the community to bring these courses to life. Next week, we will launch a kickstarter campaign to raise money to fund the creation of these courses. The three courses which will be part of the kick starter campaign are
- Computer Vision I : Introduction
- Computer Vision II : Applications
- Deep Learning with PyTorch
In addition to supporting our mission to spread AI education, the kickstarter campaign will provide you with an opportunity to register for these courses at a much discounted price.
Hands-on practical courses
We have taken a different approach while designing these courses. We will build practical applications on top of a solid understanding of underlying algorithms.
We will learn enough theoretical details to understand how algorithms work without drowning in the mathematical details.
In all three courses, our goal is to complete 3-4 practical projects from different domains. We will also have a few short assignments along with the lectures. Few of the projects we have planned are :
- Healthcare – Analyze X-ray images using Deep Learning, Build patient monitoring system
- Surveillance and security – Build a Face Recognition based attendance management system, build a people counting and surveillance system
- Autonomous vehicles – How to identify different objects on the road using the Dashcam, Lane Detection, Drowsy driver detection
- Fun applications – SnapChat and Instagram like Filters, QR Code based ID Card reader
We will go from basics to mastery, from traditional Computer Vision to advanced Deep Learning techniques, from training basic machine learning models to deploying state-of-the-art models in the cloud and building web applications. You will build a strong Computer Vision & Deep Learning portfolio and demonstrate your skills through these projects.
We will start with the absolute basics and move to mastery as we cover all the necessary algorithms and techniques to successfully complete these projects.
These are some of the libraries and frameworks that we will be working with :
For development : OpenCV, PyTorch, scikit-learn, Dlib
For deploying : Flask, ONNX and Caffe2.
The first two courses will be available in both C++ and Python. However, since most of the Deep Learning community is focussed on python, we will use PyTorch for the Deep Learning course which will be covered only in Python .
Let’s go over the material we will cover in the three courses. This list is not exhaustive, and because the courses are in the process of being created, we are open for inputs from you. Please fill out this form to send us your opinion.
Computer Vision I: Introduction
This is an introductory course for beginners. We will cover the material in both C++ and Python.
What will you learn
- Basics of OpenCV : Playing with images and videos and perform simple operations like reading, writing etc.
- Image processing and enhancement techniques like convolution, filtering, edge detection, image annotation, histograms etc.
- Image transformation techniques like color spaces, affine and perspective transforms
- Video analytics: Motion Detection, Background Subtraction, Optical Flow Estimation, and Object Tracking.
- Scene understanding using Feature Extraction and Matching
- How to build BarCode and QR Code scanners
- Introduction to modern Deep Learning techniques for image classification.
- Introduction to Linear Algebra and Statistics.
Projects
- Document Scanner
- Instagram Filters
- Smart baby monitoring system based on motion Detection and Object Tracking
- Deploy an Object Recognition Model as a Web Application using Flask
Assignments
- Lane Line Detection for Driver assist
- Human Body Measurement using a frontal/side view
- ID Card Scanner using BarCode/QRCode
Computer Vision II: Applications
In this course, we will cover many different real world applications. This builds on top of the first course. We will cover the material in both C++ and Python.
What will you learn
- Face detection, Facial Landmark Detection and its applications
- Face recognition using Eigen Faces and LBP Histograms as well as Deep Learning based Face Recognition.
- Machine Learning techniques such as SVM, Nearest Neighbors, Principal Components Analysis for solving different CV problems.
- Learn about data pre-processing, model Evaluation, hyper-parameter tuning and evaluation metrics like precision/recall, ROC Curves, IoU etc.
- Learn how to use OpenCV’s Deep Learning Module
- Learn how to extract text from images using Optical Character Recognition ( OCR )
- Case study: Object Detection using YOLO v3
Projects
- Train a 9 point Facial Landmark Detector and use it to build a Drowsiness Detector app
- Build SnapChat Filters
- Face Recognition based Home Security / Attendance management system
- Create your own dataset using the CVAT tool and use YOLO Object Detector to build a surveillance & monitoring system.
Assignments
- Style Transfer using Deep Learning
- Implement k-Nearest Neighbors algorithm
- How to curate a small dataset using CVAT Tool
Deep Learning with PyTorch
In this course, we will learn Deep Learning for computer vision using PyTorch.
What will you learn
- Basics of Neural Networks : Perceptron and Feedforward Neural Networks.
- Regression and Classification using Neural Networks
- Convolutional Neural Networks and techniques involved in training them such as Stochastic Gradient Descent, Batch Normalization, Back-propagation, data augmentation, regularization, learning rate scheduling, early stopping etc.
- Training DNNs from scratch as well as fine-tuning pre-trained models using PyTorch.
- Learn how to use visualization tools like Tensorboard and monitor the training process.
- Work on state-of-the-art models for scene understanding using Semantic Segmentation and Object Detection.
- How to optimize and export trained models for mobile using ONNX and Caffe2.
- Case Study: Human Pose Estimation using Neural Networks
Projects
- Ship Identification using drones
- Intelligent radiologist: Build an application for automatically analyzing X-Rays.
- Smart DashCam: Finding and localizing objects in a DashCam video for obstacle avoidance in autonomous driving.
- Smart Gym Trainer : Detecting bad posture during exercises like squats and plank
Assignments
- Implement algorithms like BackProp, Stochastic Gradient Descent from scratch
- Create a data augmentation class
Pricing
Our mission is to provide the best quality education in computer vision. We will support our courses with knowledgeable instructors answering questions in the forums and Industry experts to help you in your projects. That makes the creation and maintenance of the courses expensive.
At the same time, we want to make the courses affordable to people around the world. This Kickstarter campaign gives us that rare opportunity to provide deeply discounted prices for our most ardent supporters who trust us to create the best courses in Computer Vision and AI.
Do you get deeper discounts when you purchase multiple courses? Absolutely, buy the course bundles instead of individual courses!
Course Bundles
Just to recap, the three courses offered are :
- Course 1 : Computer Vision I – Introduction
- Course 2 : Computer Vision II – Applications
- Course 3 : Deep Learning with PyTorch.
We have clubbed the courses into different bundles so that you can chose the courses you are interested in as well as avail the offers. The table shows the courses included in each bundle.
Offers: Kickstarter, Bundles and Early Bird
We have given all the pricing details in the table below. Here are a few important points:
- Retail price: This the price you pay after the Kickstarter ends. Needless to say it is much higher than the Kickstarter price.
- Kickstarter price: You get 50% OFF of the retail price if you buy during the Kickstarter. Our course prices will never be as low as the Kickstarter prices.
- Bundle price: You get additional discounts when you purchase a bundle containing 2 or more courses.
It does not take a genius to figure out you will get the most bang for the buck if you purchase a bundle with an early bird offer.
Which bundle is for you?
Here is a short guide that will help you narrow down your best options
Bundle 1 : Computer Vision Enthusiast
This bundle consists of Computer Vision I: Introduction and Computer Vision II: Applications. The courses cover computer vision and machine learning concepts extensively. It also covers some basics and applications of Deep Learning, but does not go into the depth of Deep Learning techniques. It is a good option if you are not interested in training and deploying Deep Learning models, but want to get your hands on with practical computer vision projects.
Bundle 2 : Computer Vision & Deep Learning Starter
This bundle consists of Computer Vision I: Introduction and Deep Learning with PyTorch. You will learn the basics of Computer Vision and Deep Learning and apply them to really interesting projects like Human Pose Estimation, autonomous vehicles and healthcare projects. It is a good option if you are not interested in the details of Machine Learning algorithms (SVM, kNN, etc), and computer vision applications like face recognition, OCR, Snapchat filters etc. I would frankly suggest anyone considering this bundle to opt for the CV Expert Bundle instead.
Bundle 3 : Computer Vision Professional
This bundle consists of Computer Vision II: Applications and Deep Learning with PyTorch. This bundle directly starts off with advanced computer vision applications and Deep Learning. You will learn to create cool as well as impactful applications and also learn how to deploy your applications. If you are mostly interested in Deep Learning and have prior experience with OpenCV and just want to boost your portfolio with good projects, then this bundle is perfect for you.
Bundle 4 : Computer Vision Complete
This bundle consists of all the 3 courses. This bundle will take you from a beginner to a computer vision master. The courses offered will lay a strong foundation in Computer Vision, Machine Learning and Deep Learning with so many practical projects and assignments. This bundle is perfect for students who want to develop their skill set and grab a job in top Deep Learning and Computer Vision companies or start their own ventures in Computer Vision and Deep Learning.
I would like to re-iterate that the special Kickstarter price will never be available in the future after the Kickstarter ends. This is your chance to grab this super deal!
Should I take these courses?
Now that you know the curriculum and the pricing, you may be thinking if these courses are worth your time and effort.
The best curriculum
The series of courses by OpenCV.org will have the best and the most comprehensive curriculum for learning Computer Vision. Here’s why —
- Designed by industry experts: These courses are designed by a team of engineers and researchers currently working in the computer vision and machine learning industry. We are bringing all our knowledge and experience to these courses.
- Both C++ and Python versions (exclusive Kickstarter offer): You will get access to both the C++ and Python versions of the first two courses at no additional charge.
- Comprehensive & Practical: Computer vision and machine learning is vast. These courses along with our future courses will be the most comprehensive study material in computer vision available online.
Learning how to learn
Usain bolt knows a lot about running. Michael Phelps knows a lot about swimming. Yet, these elite athletes have a coach!
Learning anything new is a difficult process.
A good coach creates structured practice routines for you. They observe, guide, and motivate you. A good coach shortcuts your path to success.
Often elite athletes practice in groups. Individual practice is great, but a motivated peer group keeps you on track.
Whether you are an athlete or an engineer. The same principles of learning apply.
A lot of material covered in any book or course can be gathered from online sources and learned independently free of charge. People often start learning on their own and then lose motivation, get distracted, and stop learning.
We will provide three essential ingredients of learning — structure, guidance, and a peer group — through the following means.
- Curated projects with mentors to give you feedback on your work.
- Coding assignments that help you get a better understanding of the topics you learn.
- Course Forum monitored by several instructors where you can ask questions or discuss ideas with your peers.
- Interesting quizzes which test your knowledge just after completing a module.
- Case studies designed to boost your research aptitude.
Whistle while you work
We believe human psychology plays an important role in online learning success. Reading books or continuously watching long lecture videos is boring. We all know that.
We designed these courses to break monotony by alternatively learning and coding
- We use the amazing OpenEdX platform for creating a rich learning experience. Modules are divided into sub-modules which can be digested in short spans of 10 minutes, so that you finish more topics and don’t procrastinate.
- Online Labs with hosted notebooks ensure you can try out the code while you are going through the material.
Graduate proudly
Our primary goal is to ensure you have a rock solid foundation in computer vision because with knowledge and experience comes the confidence to take on real-world problems.
At the same time it is also our goal to make sure your resume looks great and you are seen favorably by recruiters. That is why, unlike other online courses, you will build a strong portfolio with real-world projects that will help you stand out in a competitive job market.
On successful completion of a course, you will receive a digital certificate from OpenCV.org that you can proudly display on LinkedIn
Testimonials
Our team at Big Vision LLC has vast experience in creating engaging courses. Don’t take my word for it. This is what our previous students had to say for our Computer Vision for Faces course.
Course launch dates
We have made significant progress on the course material even before the start of the Kickstarter campaign. Here are the tentative dates for release of the courses
- Computer Vision I : July 15, 2019.
- Computer Vision II : August 30, 2019
- Deep Learning with PyTorch : Oct 15, 2019.
The courses will be typically of 3-4 months duration and students will have access to the study material for 1 year lifetime! Students are free to enrol at a start date of their choosing after the courses are released.
FAQs
Let me address a few frequently asked questions
Q. Where is the link to buy?
Please use this link to go to Kickstarter project and chose a bundle which you want from the right pane.
Q. What are the benefits of supporting this Kickstarter campaign?
There are many benefits that are available exclusively for the supporters and will not be available once the campaign ends. These are:
- Heavy discounts on the courses.
- 12 months for completing the projects and getting the certification (will be 6 months after the campaign).
- Access to both the C++ and Python courses without any additional cost.
- An option to buy all future courses at a fixed price.
Q. What are the best discounts available?
- Kickstarter Discount: During this Kickstarter campaign you can get the courses at 50% discount. The campaign will end on June 13, 2019.
- Bundle Discounts: When you buy 2 or more courses as a bundle, you receive additional discounts.
Q. Do you have a discount for students?
The Kickstarter prices are deeply discounted. Unfortunately, there are no additional discounts for students. There will be a discount for students after the courses launch, but the prices will never be as low as the Kickstarter prices.
Q. Can I pay in installments?
No. Crowdfunding platforms like Kickstarter do not allow us to collect payments in installments.
Q. How long will I have access to the course?
After you start a course, it will take you 3-4 months to finish it depending on the course. See below.
- Computer Vision I : 3 months.
- Computer Vision II : 3 months.
- Deep Learning with PyTorch : 4 months.
For each course, you will have lifelong access to the course material, including all updates made to the course. In addition to the course material, students will get access to online labs during these 3-4 months. Online labs allow you to try code in hosted Jupyter notebooks while you are going through the study material.
Also, you have to complete the projects within 1 year from enrollment to receive the certificate.
Q. Will I get a certificate after taking the course? What is the criteria?
Yes, you will earn a certificate of completion from OpenCV.org. You will have to complete all projects within 1 year of enrolling in the course to receive the certificate.
Q. Why did you choose PyTorch?
PyTorch is growing fast in popularity because it is simpler and often faster than Tensorflow. We made this decision after giving it a serious thought and talking to several practitioners who have used both the frameworks.
