handtracking

This project is part of the EE 5561 course on Image Processing and Applications, during the Fall 2010 semester at the department of Electrical and Computer Engineering at the University of Minnesota Twin Cities campus. The Problem statement is to design a system to monitor the movement of hands in a video/real time webcam capture using image processing techniques like skin segmentation, haar cascade classifier and cam-shift, active contouring and other such tracking techniques. The system must be capable of taking as an input a video file and outputting the information about number of hands, x and y coordinates of the hands so identified and should thus be able to keep a log of the data. The general pipeline of the project is as shown below: Week 1: Sai Kiran Movva Introduction to the problem statement and presentation of a probable pipeline. Week 2: Amith Kamath Skin Segmentation ideas and multiband Thresholding images and results.