IMAGE SEGMENTATION: Advanced techniques for aspiring Computer Vision experts by Jeremy Cohen

IMAGE SEGMENTATION: Advanced techniques for aspiring Computer Vision experts

Master segmentation architectures and build advanced projects such as Aerial Navigation, Self-Driving Cars, or Medical Imaging.

What's included

✔️ Advanced techniques for aspiring Computer Vision experts
✔️ Real-world Portfolio project
✔️ One-on-one onboarding sessions (optional)
✔️ VIP contact and support

Would you like to understand modern Deep Learning?

Computer Vision is very broad, and not always easy to master.
Many techniques exist and can't be easily understood: there are dozens of different convolutions possible, over twenty papers relating techniques to improve the accuracy of a network, and it can be messy...

Do you know a bit about neural networks, but never went that far into understanding modern architectures?

In this course, we'll learn about image segmentation applied to drivable area detection!

Learn Image Segmentation

Learn how to use Convolutional Neural Networks to create an architecture that can segment an image and classify every pixel.

Algorithms
: 2D - Strided - Transposed - Dilated - 1x1 Convolutions, Skip-Connections, Pixel-wise classification, 5 Upsampling techniques.

Study Researcher's work

Study popular architectures that are very effective today and get better at reading research papers.

Papers:
UNet - Segnet -DeepLab - FCN

Build your own Driveable Area Detection system

Build your own image segmentation algorithm for a drivable area detection system!

To make it harder, you'll need to detect 2 classes in Paris, where there are almost no lines!

Learn, Fail, Experiment, Succeed!

🧐 Still need more precision? Look at the detailed summary here.

Master Cutting-Edge Vision Applications

You can learn to build cutting-edge Computer Vision applications such as Aerial Navigation or Driveable Area Detection.
ezgif-4-40ffc53d5347.gif 16.5 MB

You'll know how to adapt your projects to any situation, and how to use this in the real world.
Enroll, and get closer to Computer Vision expertise.
Enroll Now

Success Story & Testimonial

"I followed the segmentation course. It's well structured, and we have every resource necessary to get results and even adapt it to other problems. "
Screenshot 2020-07-07 at 15.12.07.png 1.1 MB


Staberlin did an amazing job in his project and developed a drivable area detection system using SegNet architecture and solved the Paris Challenge.

Look at his video results here!
Staberlin Valentain, AI Researcher Intern
Enroll Now

FAQs

What are the prerequisites?

Coding in Python, understanding neural networks optimization, and handling Keras library are the main prerequisites.
If you've never done all of this, it will not be possible to get the results I talked to you about.
If you do, there is no reason you shouldn't succeed.

Can I join if I don't have access to a GPU?

Yes, the course will work with Google Colab. It's free and you can use the provided GPU. The installation will be as easy as the click of a button.

I already know it, will I learn something?

If you already know semantic segmentation, you might learn some things but rapidly arrive to the challenges.
The course has 2 challenges: Drive in Paris, where there are no lines, and drive in Costa Rica, with lots of shadows and sun.

Image Segmentation is getting more and more useful. The number of applications is limitless and the potential is amazing. Start today, take your skills to the next level.

Enroll Now

Jeremy Cohen

Artificial Intelligence & Self-Driving Car Engineer, Head Dean of France School of AI, and Machine Learning Lecturer.
I started thinkautonomous.ai to help aspiring AI & Self-Driving Car Engineers to land their dream job. Working in the industry of the future requires skills and passion. You can build your skills here, where you'll create relevant projects that are used every day in autonomous robots & AI engines.

👉 Learn more