Become a Computer Vision Engineer
Work with images and video, from OpenCV to vision transformers.
10 lessons, in order
The route
Read these top to bottom. Nothing here assumes knowledge you have not been given earlier in the list.
-
Python for AI
Python basics
Python is a way of writing step-by-step instructions for a computer using words close to English. It is the language almost all AI is built in.
-
Python for AI
NumPy
NumPy lets you do one operation to millions of numbers at once instead of one at a time. It is the foundation every AI library in Python is built on.
-
Computer Vision
How images are stored in a computer
An image is a grid of numbers, one small set of numbers per dot, and every single thing in computer vision is arithmetic done on that grid.
-
Computer Vision
OpenCV
OpenCV is the free toolbox that reads, reshapes and measures images for you, so you never write resize or blur or edge detection by hand.
-
Deep Learning
Convolutional neural networks (CNN)
A CNN looks at a small patch of an image at a time and slides the same pattern detector across the whole picture, which makes vision models small…
-
Computer Vision
Image classification
Image classification is giving a whole picture one label from a fixed list of choices, which is the first real task most vision projects start with.
-
Computer Vision
Object detection
Object detection finds every object in a picture and draws a box around each one, so you get what it is and where it is, not only what is present.
-
Computer Vision
YOLO
YOLO looks at the whole picture in one pass and reports every object at once, which is what made object detection fast enough for live video.
-
Computer Vision
Image segmentation
Segmentation labels every single dot of a picture, so instead of a rectangle around an object you get its exact outline.
-
Computer Vision
Vision transformers
A vision transformer cuts a picture into squares and lets every square look directly at every other square, instead of sliding a small window across…
When you finish
Build something with it. Reading alone will not make it stick — the first time you debug your own model is the moment it becomes real. Try the projects, and keep the error database open in another tab.