Computer Vision, CS378/395T, Fall 2007 Outline of topics • Image formation – Perspective, orthographic projection properties, equations, effects – Pinhole cameras – Thin lens – Field of view, depth of field • Color – BRDF – Spectral power distribution – Color mixing – Color matching – Color spaces – Human perception • Binary image analysis – Histograms and thresholding – Connected components – Morphological operators – Region properties and invariance – Distance transform, Chamfer distance • Filters – Application/effects of – Convolution properties – Noise models – Mean, median, Gaussian, derivative filters – Separability • Edges, pyramids, sampling – Image gradients – Effects of noise – Derivative of Gaussian, Laplacian filters – Canny edge detection – Corner detection – Sampling and aliasing – Pyramids – construction and applications • Texture – Analysis vs. synthesis – Representations • Grouping – Gestalt principles – Clustering: agglomerative, k‐means, mean shift, graph‐based – Graphs and affinity matrices • Fitting – Hough transform – Generalized Hough transform – Least squares – Incremental line fitting, k‐means – Robust fitting: RANSAC, M‐estimators – Deformable contours, energy functions • Multi‐view geometry – Human stereopsis and disparity – Geometry of two views: stereo rigs – Case of calibrated cameras and parallel optical axes – Epipolar geometry and the epipolar constraint – Triangulation – Essential matrix • Stereo reconstruction – Rectification – Non‐geometric constraints for correspondences – Dense vs. sparse stereo matching • Camera calibration, self‐calibration – Intrinsic parameters – Linear perspective projection equations – Estimating the projection matrix – Fundamental matrix – Robust computation for uncalibrated views • Local invariant features – Classes of transformations – General interest operators – Scale invariant detection, scale‐space – DoG, SIFT detection of keypoints – Affine invariant detection – Local descriptors – Application for wide baseline stereo • Indexing local features – Search task – Bags‐of‐words representation, computing visual vocabularies – Inverted file indexing • Model‐based recognition – Interpretation trees – Alignment, pose consistency – Pose clustering, voting – Verification • Learning and supervised classification – Generative vs. discriminative models – Bayesian inference – Support Vector Machines – Boosting, Adaboost – Nearest neighbors – Cascade of classifiers • Object recognition examples – Eigenfaces – Viola‐Jones face detection, rectangular features – Models of shape and appearance – Part‐based model, learning with weak supervision • Motion – Motion field equations – Motion parallax – Optical flow – Aperture problem and brightness constancy – Lucas‐Kanade optical flow algorithm – Discrete matching algorithm – Coarse‐to‐fine computation – Image warping via flow fields • Tracking – Recursive estimation – Hidden states and measurements – Tracking as inference – Linear dynamics models – Kalman filter – Data association – Particle filters, Condensation algorithm – Example‐based pose estimation+motion graphs
View Full Document