# Topics - Introduction to computer vision - Cameras and image formation - Simple edge detection - Linear filtering and convolution - Gradient filters and blurring - Image pyramids: Gaussian and Laplacian - Nonlinear filtering (bilateral filtering) - Frequency domain and Fourier transforms - Image bases and Fourier analysis - Machine-learning fundamentals for vision - Nearest-neighbor and linear regression - Overfitting and generalization - Linear classifiers and logistic regression - Stochastic gradient descent - Neural networks: nonlinearities, architectures, regularization - Optimization and backpropagation - Computation graphs, momentum, autodiff - Convolutional neural networks - Convolution layers, pooling, normalization - Scene understanding and semantic segmentation - Object detection - Sliding-window and region-based CNNs - Instance segmentation (Mask R-CNN) - Image synthesis - Texture synthesis - Generative adversarial networks (GANs, conditional GANs) - VQ-VAEs and diffusion models - Temporal models - 3-D convolutions, recurrent nets, LSTMs - Representation learning - Transfer learning, autoencoders, self-supervised learning - Language and vision - Vision–language models (e.g., CLIP) - Sound and touch - Multimodal self-supervision - Image formation revisited: camera models, projection, plenoptic function - Multi-view geometry - Fitting geometric models (e.g., RANSAC) - Structure from motion and stereo - Motion estimation and optical flow - Aperture problem, keypoint tracking - Light, shading, and color - Shape-from-shading, intrinsic images, color perception - Embodied vision - Learning from demonstration, reinforcement learning for vision - Recent network architectures - Implicit representations, Vision Transformers, NeRFs - Ethics, bias, and disinformation in computer vision - Dataset bias, algorithmic fairness, image forensics