Computer Vision
Prerequisites: Foundation courses.
Recommended path
- Everyone should go through CS231n. The lectures and notes are unparalleled in terms of introductory material on Deep Learning and Computer Vision. The 2016 version is more outdated, but the lectures up to backpropagation are so well explained by Karpathy that we'd recommend going through the first four 2016 lectures and then switching to the more updated 2017 version.
- For a very similar course to CS231n but a more updated version, take a look at the UMichigan course, which is taught by one of CS231n's main instructors (Justin Johnson).
- In addition to these deep learning focused courses, UW's Ancient Secrets of Computer Vision (by one of the YOLO authors) is a lot broader and goes through topics such as SIFT and HOG. So, if you want to learn about CV outside of Deep Learning, the first half of this course is your chance.
Course | Year | Description | Difficulty Level | Resources |
2019 | This course is essentially an updated version of CS231n. It includes a lot of overlapping topics and features new content such as Transformers and 3D Vision | Medium | ||
2018 | The most well-rounded computer vision course as it not only teaches you the deep learning side of CV but also older methods like SIFT and optical flow. This course is very complementary to CS231n. | Medium | ||
2016, 2017 | This is one of the most popular deep learning courses online, and for a good reason. It builds up the foundations of deep learning via computer vision examples. Beyond the lectures, CS231n also has some of the best complementary course material including notes, assignments and projects. | Medium |
Resource | Year | Description | Type |
2020 | This course is great for anyone who has already taken an intro CV or DL course and wants to explore ideas like neural rendering, interpretability and GANs further. | 🎥 | |
Continuously updated | Distill's interactive articles on feature visualizations and circuits are an amazing resource to improve your understanding of neural network layers and neurons. | 🌐 | |
Continuously updated | A large percentage of Kaggle competitions is focused on computer vision tasks. Reading through the discussion forums and Kernels will give you an excellent overview of how recent CV methods are being applied in practice. | 🤖 |
Last modified 2yr ago