Though process capture

I sincerely urge my friends / authors to start capturing the thought process by means of comments to this post.

Crude approach

Well, since not too much in to how to get it done here is some crude way to get started

[1] Video capture
[2] captured Data processing / Decoding
Should look through what kind of decoded data is needed for the algorithm to be implemented purely dependent on the [3]
[3] Applying algorithms (Major Task)

Comments and modifications are welcomed by the authors

Synopsis

SYNOPSIS

Title: Vision Based Hand Gesture Recognition For Intelligent Human Computer Interaction.

Goal: To develop a tool which uses real time hand tracking for interacting with computer.

Introduction:

A Hand gesture is defined as a dynamic movement,such as waving “good bye..”.Gestures are made in 2 ways. First,a simple posture of hand and change in the position or orientation of the hand. Second,Position and orientation is not changed ,But moving fingers in some way. Some complex gestures may include both of the above.

First step is to collect the raw data. And this raw data is analyzed by different algorithms to find out some pattern out of it and further it is used in hand gesture recognition. We use cameras to collect raw data.

Once the raw data has been collected it must be analysed to determine if there are any gestures recognized. There are several algorithms available.

Approach:-

Hidden Markov Models:

HMM is defined as a set of states of which one state is starting state , a set of output states and some state transitions. Each state transition is represented by state from which the transition starts,the state to which transition moves ,the out put generated and the probability that the transition is taken. In our context each state could represent a set of possible hand postures. The state transitions represent the probability that the certain hand position is changed into other. The corresponding output represents a

specific hand posture and a sequence of postures represent hand gesture. We can a group of HMMs ,one for each gesture and runs a sequence of input data through each HMM. The HMM with highest forward probability determines the most likely hand gesture.

Advantages Of using HMMs:

1.Can recognize large set of postures and gestures.

2.With adequate training, high accuracy can be achieved.

3.Well discussed in the Liturature.

Steps:

1. To select which approach to be used for recognition.

2. To select the tools to be used.

3. Gathering training data and start prototype.

4. Analysing the data which collected in the above step.

5. Some classification of the data based on algorithms.

6. Real time video tracking.

7. Motion Detection using frame by frame analysis .

8. Applying HMM for gesture recognition.

9. Testing

Tools :MATLAB and C++

LATEST PAPERS:

[1] Qing Chen1, Marius D. Cordea1, Emil M. Petriu1, Thomas E. Whale,”Hand-Gesture and Facial-Expression Human-Computer Interfaces for

Intelligent Space Applications

”,University of Ottawa, ON, Canada, CRC, Ottawa, ON, Canada,Budapest Tech Polytechnic Institution, Hungary, Budapest University of Technology and Economics, Hungary

,MeMeA 2008 – IEEE International

Workshop on Medical Measurements and Applications

Ottawa, Ontario, Canada - May 9-10, 2008

.

[2]Reza Hassanpour, Stephan Wong, Asadollah Shahbahrami, ”Vision­Based Hand Gesture Recognition for Human Computer

Interaction: A Review

”,IADIS International Conference Interfaces and Human Computer Interaction 2008.

[3]Qing Chen, Marius D. Cordea

and Emil M. Petriu,”Human–Computer interaction for smart

environment applications using hand gestures

and facial expressions

”,University of Ottawa, 800 King Edward, Ottawa,

Ontario, K1N 6N5, Canada

,Int. J. Advanced Media and Communication, Vol. 3, Nos. 1/2, 2009.

[4] Joseph J. LavioLa Jr ,”A survey of Hand Posture and gesture recognition Techniques and Terminalogy.”,Department of Computer Science ,Brown University ,Providence Rhode Island,02912.

[5]C. Keskin, A. Erkan, L. Akarun,”real time hand tracking and 3d gesture recognition for

interactive interfaces using hmm

”, Computer Engineering Dept.

Boğaziçi University

,E-Mail: ckeskin@ttnet.net.tr, {erkanays,akarun}@boun.edu.t

r

[6] Nguyen Dang Binh, Enokida Shuichi, Toshiaki Ejima,”Real-Time Hand Tracking and Gesture Recognition System”, Nguyen Dang Binh, Enokida Shuichi, Toshiaki Ejima

Intelligence Media Laboratory, Kyushu Institute of Technology

680-4, Kawazu, Iizuka, Fukuoka 820, JAPAN

[ndbinh, enokida, toshi]@mickey.ai.kyutech.ac.jp

, GVIP 05 Conference, 19-21 December 2005, CICC, Cairo, Egypt.

Vague Dead lines

Experience says it had always been to difficult to draw lines in the air and determine the ends of the purpose. So i am puting forth limits on how much should be accomplished till what time in a vague fashion, the other authors of the blog would also revise the deadlines if needed.

The Project would mainly deal with Hand Gesture Recogntion.

A minimum/maximum of 5 day gap b/w each dead line.
Midsem work should be completed before Sep 19th


Aug 10th - Submission of Synopsis and short write up about the project and goals
Aug 15th - Complexities and complications in Hand gesture Recognition, Algorithms involved in implementing the Gesture Recognition
Aug 20th - Short write up on Language of implementation and on how-to
Aug 25th - Work Division
-------------10 days-------------
Sep 5th - Working modules/prototypes of implemented work
Sep 10th - Proceed on development process
Sep 15th - Short write up on work done so far while still carrying on with the prj
Sep 20th - Presentation and report of work done so far
--------------EXAMS -------------
Sep 30 th - Revisitng the Prototypes and continuing Development
Oct 5 th - Improving the work and collabarting the work done by all the three
Oct 10th - Test cases and analyzing the performance
-------------15 Days --------------
Oct 25 th - Development
--------------Yet To Be Decided ----------------
Nov 20th - Dead Line Of Project.

Nov 26 th - Exams

Gesture Recognition

This is a totally unofficially blog to host the proceedings and keep track of progress of our Semester project work on Gesture Recognition. For which Naresh, Anil and me are going to contribute.

Note: to the authors if at all they modify some posts
Enter the modified text in Bold and Italic using the edit options.