Computer Vision
Code | 227 |
---|---|
Semester | Spring |
Class Hours - Lab Hours | 3 - 2 |
Lecturers | Petros Maragos |
Description
Introduction to the theory of fundamental problems in computer vision, synopsis of evidences from biological vision, mathematical models and computational algorithms for their solution, and description of selected applications. Visual sensors, perspective, and image formation. Photometry. Color. Spatio-temporal processing of visual signals: Multidimensional linear filters and Fourier/Gabor analysis. Morphological operators and nonlinear filters. Multiscale image analysis and pyramids with linear (Gaussian scale-space) and nonlinear methods (geometric diffusion). Detection of edges and other geometric features, and extraction of descriptors. Analysis and modeling of shape and texture. Visual motion estimation. Multiview geometry and stereopsis. Curve/surface evolution, active contours, and levelsets. Image segmentation: geometric, statistical and graph-theoretic methods. 3D reconstruction. Recognition of objects and actions by combining computer vision and machine learning methods. Selected applications in biomedicine, robotics, digital arts, and internet.