Computer Vision Study Plan

What is computer vision? A brief history, book overview, and notation.

Geometric primitives, transformations, photometric image formation, and digital camera concepts.

Point operators, linear filtering, Fourier transforms, pyramids, wavelets, and geometric transformations.

Least squares fitting, RANSAC for robust estimation, and total variation regularization.

Neural network fundamentals, backpropagation, CNNs, and modern architectures like ResNet and Transformers.

Instance recognition, image classification, object detection, and semantic segmentation.

Points, patches, edges, contours, lines, vanishing points, and segmentation.

Pairwise alignment, image stitching, global alignment, and compositing.

Translational alignment, parametric motion, optical flow, and layered motion.

HDR imaging, super-resolution, denoising, matting, and texture synthesis.

Camera calibration, pose estimation, SfM, and simultaneous localization and mapping.

Epipolar geometry, stereo matching, multi-view stereo, and monocular depth.

Shape from X, 3D scanning, surface representations, and model-based reconstruction.

View interpolation, light fields, video-based rendering, and neural rendering.