Introduction to computer vision concepts: Review

Page 7 of 9, the “Use computer vision tools” unit, had no tools. It was just a wall of text describing things I could do *if* I had the tools. I read about image classification, object detection, and segmentation, looked at some stock photos, and then the unit just ends. No sandbox, no API call, not even a pre-written script to run. It feels like reading the back of a video game box instead of playing the game.

The whole module is just a high-level glossary. You'll learn the difference between a CNN and a Vision Transformer by reading a paragraph about each. You'll see what image generation is by looking at a generated picture of a cat in a hat. It's all passive. For a module with "computer vision" in the title, I expected to see a computer do some vision. The knowledge check only tests if you remember definitions from the previous pages. I blew through it in under 30 minutes, most of which was just clicking “Continue.”

I guess it’s fine if you're a program manager who needs the buzzwords for a meeting. But as a student trying to actually build something or prep for an interview, this is basically useless. It's a map that shows you the names of the continents without giving you a boat.