Project Overview

Ambient Silhouettes: From Vitrivius' proportions to calm displays

Faculty Sponsor

Elodie Fourquet (efourquet@colgate.edu)

Department(s)

Computer Science

Abstract

The goal of this project is to develop a camera tracking application so that human gestures control a display. This project requires the design and evaluation of computer vision algorithms to track and recognize a moving human in a video stream. Using the temporal and spatial coherences in mapping body part proportions (as specified by Vitruvius/Leonardo ideal man canon) within and between video frames, represent a new approach. During this summer one student experienced with OpenCV (the computer vision library) will be responsible to develop further the code, including the design of a rigid test suite to facilitate the calibration of multiple parameters of the application. The other student responsibilities will be to review the literature so as to measure and compare our new approach against the known state-of-the-art methods and to write-up a report synthesizing our approach and the result of its evaluation.

Student Qualifications

COSC 301 programming experience is required for these positions. Experience with OpenCV is a plus for one of the position. Writing, synthesis and/or analytical research skills is required for the other position (this ability isn't limited to an experience in the natural science; an experience within the humanities is perfectly adequate).

Number of Student Researchers

2 students

Project Length

10 weeks


Applications open on 01/15/2017 and close on 02/07/2017


<< Back to List





If you have questions, please contact Karyn Belanger (kgbelanger@colgate.edu).