Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Elsevier PPT
1. Authors Presented By
Ram Pratap Sharma, Ram Pratap Sharma
Asst. Prof. Gyanendra Kr. Verma
Department Of Computer Engineering
National Institute Of Technology, Kurukshetra
1/27/2016 Presented By: Ram Pratap Sharma
1
3. In this modern world, computers are being used by a large number of
people and its demand is still growing day by day
Over the years many devices are being designed to make an easy
interaction between human and computers
The increase in human–computer interaction has made user interface
technology more important
The most natural and intuitive alternative to the cumbersome devices (like
keyboard, mouse etc.) is to use human hand gesture for human – computer
interaction
A gesture can be defined as a physical movement of the hands, arms, face
and body with the intent to convey information or meaning
It can also be defined as a movement of a limb or the body as an expression
of thought or feeling. (Oxford Concise Dictionary1995)
1/27/2016 Presented By: Ram Pratap Sharma 3
4. 1/27/2016 Presented By: Ram Pratap Sharma 4
To design a simple, natural, precise and robust
framework i.e. a computer vision algorithm for Human
Computer interaction using human hand gesture
5. There are various bodily motion which can originate gesture but the
common form of gesture origination comes from the face and hands
Contact based and vision based technologies are two main types of
technologies used for robust, accurate and reliable hand gesture recognition
systems
In [4], a method for detecting finger from the detected hand, can be used as
a non-contact mouse, has been proposed
In [5], the author has used Lucas Kanade Pyramidical Optical Flow
algorithm to detect moving hand and K-means algorithm to find center of
moving hand
In [6], a fast, simple and effective gesture recognition algorithm for robot
application has been presented which automatically recognizes a limited set
of gestures
1/27/2016 Presented By: Ram Pratap Sharma 5
6. In [7], a comparative analysis of different segmentation techniques and
how to select an appropriate segmentation method for the system have
been presented
1/27/2016 Presented By: Ram Pratap Sharma 6
8. Recorded each video stream of duration time approximately 10 seconds at
the rate of 30 frames per seconds and at the resolution of 1280x720 using
digital camera of 8 megapixel
Experiments performed in three different sessions, each having 6 different
class of gesture (total images for each session = 50x6)
All the experiments were carried out in MATLAB 8.1.0 (R2013a), on a 64-
bit Intel Pentium processor (2.40 GHz) with 2 GB RAM
1/27/2016 Presented By: Ram Pratap Sharma 8
9. Session 1 Session 2 Session 2
After performing the experiments we have seen that the overall accuracy of
the system is 95.44%
The minimum accuracy we have achieved by class 3 gesture in session 3
due to the gesture shapes and positions
1/27/2016 Presented By: Ram Pratap Sharma 9
10. Gesture recognition is very challenging and interesting task in terms of
accuracy and usefulness in computer vision
Rotation, illumination change, background variations, and pose variation
of hand makes the problem more challenging
Most important advantage is that we can efficiently interact with the
application from a distance without any physical restriction
We have proposed an algorithm for recognizing the hand gestures in a
constant background and good lighting conditions
This application can be very helpful for the people of developing countries
and most importantly to physically challenged user
After performing the experiment we have achieved overall accuracy of
approx. 95.44% which is a quite good result after comparing with other
systems
1/27/2016 Presented By: Ram Pratap Sharma 10
11. We can use some other mode of communications (like speech, head, face
etc.) with hand together to acquire more accurate results and number of
gestures
We need to build more robust algorithm for both recognition and detection
even in the cluttered background and a normal lighting condition
We need to extend the system for some more classes of gesture as we have
implemented it for only 6 classes of gesture
Also the vision-based approach has preferred to be less complex than
3D model based approach. But for the future perspective we should also
have to make the progress in 3D representation of gesture
1/27/2016 Presented By: Ram Pratap Sharma 11
12. [1] Rautaray S S and Agrawal A, “Vision based hand gesture recognition for human
computer interaction: A survey”, Springer Transaction on Artificial Intelligence
Review, pp. 1-54, 2012.
[2] Payeur P, Pasca C, Cretu A, and Petriu E M, “Intelligent Haptic Sensor System for
Robotic Manipulation”, IEEE Transaction on Instrumentation and Measurement,
54(4), pp. 1583-1592, 2005.
[3] Hasan M M, and Mishra P K, “Hand Gesture Modeling and Recognition using
Geometric Features: A Review”, Canadian Journal on Image Processing and
Computer Vision, 3(1), pp. 12-26, 2012.
[4] Kang S K, Nam M Y, and Rhee P K, “Color Based Hand and Finger Detection
Technology for User Interaction”, IEEE International Conference on Convergence and
Hybrid Information Technology, pp. 229-236, 2008.
[5] Rautaray S S and Agrawal A, “A Novel Human Computer Interface Based On
Hand Gesture Recognition Using Computer Vision Techniques,” In Proceedings of
ACM IITM’10, pp. 292-296, 2010.
1/27/2016 Presented By: Ram Pratap Sharma 12
13. [6] Malima A, Özgür E, and Çetin M, “A Fast Algorithm For Vision-Based Hand
Gesture Recognition For Robot Control”, IEEE Signal Processing and
Communications Applications, pp. 1-4, 2006.
[7] Ibraheem N A, Khan R Z, and Hasan M M, “Comparative Study of Skin Color
based Segmentation Techniques”, International Journal of Applied Information
Systems (IJAIS), 5(10), pp. 24-38, 2013.
1/27/2016 Presented By: Ram Pratap Sharma 13