I took Stanford's CS231N - Deep Learning for Computer Vision, in Spring 2025. I thoroughly enjoyed the course and found myself going back to the slides to review things months after the course ended as well. Each time I did this, I learned something I had missed before, or clarified ideas I had misunderstood previously.
While this phenomenon might be true of any class one takes where course material is of such high quality, I found it to be exceptionally true in case of CS231N. Since I took the course, Stanford made the lecture videos available on YouTube. Having such easy access to videos makes it possible to re-watch lectures I rushed through or missed before, and slow down and really dig into topics of interest. I am grateful to Stanford and the course staff for making this material publicly available.
But I have found myself wanting a simpler summary of these course notes - nothing more, nothing less - that was a bit easier to navigate than having to search through slides or scrub through videos each time I wanted to refer back to something.
So here is my own course summary, written months after having taken the course and after re-watching some of the 2025 lectures posted on YouTube and reading through course slides. I have prepared notes in markdown for a few of the initial lectures, and I do so at my own pace, as I revisit course material, sometimes discussing it with a group of friends in a small Sunday study group.
These notes started as a summary for my own benefit. If they are useful to anyone else, that is a huge bonus. There is no time-table to complete notes for all of the CS231N lectures, though I hope to write a light summary for most of the 2025 CS231N lectures. Markdown has its limitations and formatting on my local machine vs github can vary sometimes. Someday I might go back and rewrite these notes in latex. For now, I am prioritizing reviewing and summarizing the material with quick markdown notes, over typesetting detailed latex notes since I tend to get carried away with latex formatting quirks.
Any mistakes in my notes are my own. Corrections are most welcome. For authoritative and comprehensive study material, please refer back to the original course materials or many of the resources referenced in the slides and course notes.
Resources
- 2025 Course website including slides and course notes
- 2025 Lecture Videos (Playlist): https://www.youtube.com/playlist?list=PLoROMvodv4rOmsNzYBMe0gJY2XS8AQg16
Organization of the course
- Lecture 1 : Intro
- Lectures 2-4 : Deep Learning Basics
- Lectures 5-12 : Perceiving and Understanding the Visual World
- Lectures 13-16 : Generative and Interactive Visual Intelligence
- Lectures 17-18 : Human-Centered Applications and Implications