Template Image
Computer Vision
Offered in Fall 2005, the course covered camera caliberation, lighting, shading, segmentation and shape analysis. The course emphasized on use of Mathematics in the domain of Differential Geometry, Linear Algebra, Statistics and Probability. The course was evenly balanced and had a couple of programming assignments apart from normal examinations. Books on Robot Vision by B.K.P Horn and Computer Vision - A Modern Apporach by Forsyth & Ponce were used as texts.
Programming assignments were done in C and MATLAB. Following are list of mini-projects implemented in class. Each project has a description, source code, screen shots and a working demo.
Camera Calibration
Given a point in 3D space and its corresponding images at two cameras. The goal was to determine extrinsic camera parameters. See here for more. Description.
Contour Tracking
To determine recognize an object and its shape in a region of interest using contour tracking. See here for more.Description.
Image Identification & Search Engine
A term project to design an image search engine where an image is used to query database. Results display images ranked according to similarity metric. See more forDescription.
Image smoothing filters
To implement common image filter algorithms to run on stream processor. See here to know more about the project Description.
Single View Reconstruction
Paper implementation of above. See here to know more on this Description.
Image Matting
Paper implementation of above. See here to know more on this Description.
Seamless Tiling
Paper implementation of above. See here to know more on this Description.
External Mesh Deformation
Paper implementation of above. See here to know more on this Description.
Visual Tracking - Master thesis
Topic was Stream Processor Based Real-Time Visual Tracking using Apperance Based Approach. See here to know more on this Description.