Development of a robust video tracking system based on kinect.
Tran, Hoai Nam.
Date of Issue2012
School of Electrical and Electronic Engineering
Recently, there has been a proliferation of 3D audio technologies, it has become a challenging task to research and developed new ideas to help out in this area. As a part of a development project in 3D sound systems, one of the most challenging tasks is to locate the position of human ears so that the sound systems could accurately point the sound beam to the ears. As a result, a numerous number of sound effects can be created. This project serves as an important addition to the 3D Audio System by building and developing a program to track the location of the ears of human when the body is moving around and thereby making it possible to beam accurately audio sound to the ears. As a result, the developers can play around with this application to create many sound effect applications. The main task of the author is to build an Ear Tracking Program (ETP) using a sensor camera Kinect- introduced by Microsoft recently, a PC system on Windows 7 operating system with great processing capacity. The ETP was written in C# programming language on Visual Studio 2010 environment, making used of an OpenCV library and synchronize with Kinect Software Development Kit (SDK). The ETP takes in a colour live video stream from the Kinect and run the Visual Studio codes on it. The codes will analyse the live video and track the relative locations of two ears with 2 methods – which are presented in the later part of this report. The locations of two ears are indicated in a blank dark window and the corresponding position data of these locations will be transferred over to the next stage –3D Audio Beaming stage. Due to restricted time and resources, in this report, the author mainly focuses on the Ear Tracking stage and all the matters regard to. The methods and algorithm of ETP are fully described in this Final Year Project report, together with the results and discussions. Some recommendations for further implementations are also discussed.
DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Final Year Project (FYP)
Nanyang Technological University