Manoj kumar RAJAGOPAL

Bâtiment C-310
Département of EPH
# 9 , Rue Charles Fourier
91000 Evry
France
manojkumar.rg@gmail.com
Phone: +33 663 824 724

My CV

Research Interests

My research interests are Human-Computer Interaction, Computer Vision and Machine learning. In Human Computer Interaction I like to bring expressivity of a real human in a virtual human. I am also interested in computer vision and machine learning problems related to optical flow, gesture and activity recognition, and dimensionality reduction techniques.

About Me

I finished my PhD at Telecom Sudparis working under the supervision of Patrick Horain and Catherine Pelachaud of CNRS, Telecom ParisTech. Prior to this I worked as a Project Associate in IITK , India for one year and 2 years as a Lecturer in leading Universities in Tamil Nadu, India. I completed my Bachelor of Engineering in VIT, Vellore and Master of Engineering in CIT, Coimbatore, India.

PhD Thesis

Manoj kumar Rajagopal, "Cloning with Gesture Expresivity", PhD thesis 2012TELE0013 at Telecom Sudparis defended on 11 May 2012.

Publications

1. M. K. Rajagopal, P. Horain, C. Pelachaud,  " Animating a conversational agent with user expressivity " Proceedings in in 11th International Conference on Intelligent Virtual Agents (IVA 2011), Reykjavik, Iceland, September 15-17, 2011, Vol.6895, Pages 464-465.PDF

2. M. K. Rajagopal, P. Horain, C. Pelachaud, " Virtual Cloning Real Human with Motion Style " Proceedings in 3 rd International Conference on Intelligent Human Computer Interaction (IHCI 2011) , Prague , Czech Republic, August 29-31, 2011.PDF

3. D. A. Gómez Jáuregui, P. Horain, M. K. Rajagopal, S. S. Kumar Karri, " Real-Time Particle Filtering with Heuristics for 3D Motion Capture by Monocular Vision ", Proceedings of the 2010 IEEE International Workshop on Multimedia Signal Processing (MMSP'10) , Saint-Malo, France, October 4-6, 2010.PDF

4. D. Glowinski, M. Mancas, P. Brunet, F. Cavallero, C. Machy, P. J. Maes, S. Passchalidou, M. K. Rajagopal, S. Schibeci, L. Vincze, and G. Volpe, " Toward a model of computational attention based on expressive behavior: applications to cultural heritage scenarios ", A. Camurri, M. Mancini, G. Volpe (Eds.), 2010. Proceedings of the 5th International Summer Workshop on Multimodal Interfaces (eNTERFACE'09), DIST-University of Genova, Genova, Italy. ISBN: 13-978-88- 901344-7-0. Pages 71-78.PDF

5. M.Mancas, D.Glowinski, P.Brunet, F.Caveller, C.Mach, P-j.Maes, S.Paschalidou, M.K.Rajagopal ,S.Schibeci, L.Vincze, G.Volpe, " Hypersocial museum: addressing the social interaction challenge with museumscenarios and attention-based approaches ", QPSR of the numediart research program, Vol. 2, No. 3 , September 2009, pages 91-96..PDF

Earlier Projects


Projects Demo
E-SignBoard: The aim of this project is to make a signboard that can be controlled by pointing gestures. It is designed such that just by pointing the finger towards the screen, that portion of the board will get highlighted and we will get complete information about the matter contained in that part. Image of the hand is captured using webcam and it is converted in to binary image after it converted in to greyscale image. The binary image is chopped off to to obtain the region containing the hand only. The chopped region is divided in to fixed number of zones and the line joining the centre of mass of the first and the last zone is found out. This line which indicates the direction of hand is used to determine the region of the screen which the user is pointing at. The cursor moves accordingly, following the motion of the hand. When the cursor stays in a particular region for a specified time interval, entire information pertaining to that region is displayed on the screen for the next few seconds.
Optical Flow: The objective of this project is to implement the Lucas kanade optical flow algorithm to track the pixel from one frame to the next frame. This is a first step towards video stabilization. Starting point of the arrow mark is the present frame pixel and the ending point of the arrow mark is the pixel corresponding to previous frame.
Image Segmentation using Level Set Method:Image Segmentation is a process which is used to find the object in the given picture. I have implemented Level set Method for this concept. U Using this level set method I am drawing a contour at the starting stage. Then we are preceding further to find the border of the image present in the picture. Experiment results shows that this approach can use a large time step to speed up the segmentation process and achieve good results on real images. This is implemented in C++ Language using OpenCV.
   
Note
I am conducting a survey for my research work based on my results. You are invited to give your views in the survey Click Here

Hit Counter by Digits This is a Webcounter