Stanford AI Course Map 中文 GitHub ↗
← Back to the course map
Stanford · 2026—27 CS 231N Foundations ★★★★☆

Core course · Foundations

Deep Learning for Computer Vision

中文:计算机视觉的深度学习

The best-known deep-learning course in computer vision: from kNN to Transformers, with assignments that make you hand-write backpropagation and train CNNs.

Draft translation: official facts are checked; wording may change after native review.

3–4 units Spring 2027 Letter or Credit/No Credit ★★★★☆
CVDLVisionProject

Course positioning

CS 231N is the standard entry point to visual recognition and emphasizes the full implement–train–debug loop. Its assignment design (A1–A3 plus a final project) has been borrowed by many courses, and for many people it is the first place a neural network actually runs.

What it covers

  1. Image classification: kNN, linear classifiers, losses and optimization
  2. Neural networks: backpropagation, multilayer networks, activations
  3. CNN architectures: AlexNet, VGG, ResNet, efficient networks
  4. Training engineering: initialization, normalization, regularization, learning rates, data augmentation
  5. Visualization and interpretability: saliency maps, feature visualization, adversarial examples
  6. Detection and segmentation: R-CNN family, YOLO, semantic/instance segmentation
  7. Frontier topics: video, generative models, multimodality, Transformers
  8. Final project: usually a real vision task

Prerequisites

Fluency in Python/NumPy; calculus and linear algebra (MATH 19/51); basic probability and statistics (CS 109 level).

Assessment & workload

Three programming assignments plus a final project; the workload is heavy but worth it. Check the current syllabus for details.

Who it's for

People aiming at CV, multimodality, robot perception, or autonomous driving; engineers who want the low-level implementation skills behind deep learning; prospective vision PhD students.

Study advice

Hand-write BatchNorm in A2 and the Transformer in A3 instead of copying. For the final project, don't just chase SOTA — a small, well-explained error analysis is more convincing.

Related courses

Official course description

Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification and object detection. Recent developments in neural network approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of neural-network based deep learning methods for computer vision. During this course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. We will cover learning algorithms, neural network architectures, and practical engineering tricks for training and fine-tuning networks for visual recognition tasks.Prerequisites: Proficiency in Python - All class assignments will be in Python (and use numpy) (we provide a tutorial here for those who aren't as familiar with Python). If you have a lot of programming experience but in a different language (e.g. C/C++/Matlab/Javascript) you will probably be fine.College Calculus, Linear Algebra (e.g. MATH 19, MATH 51) -You should be comfortable taking derivatives and understanding matrix vector operations and notation. Basic Probability and Statistics (e.g. CS 109 or other stats course) -You should know basics of probabilities, gaussian distributions, mean, standard deviation, etc.