Sheetal Lahabar

Present Address Permanent Address

****, ****** **, ****** ****, 74, Patel Nagar

New Delhi - 110070 Gorewada, Gittikhadan Road

Email: gia089@r.postjobfree.com Nagpur - 13

Phone: +919*********

Objective To obtain a position that will enable me to use my strong programming oganizational skills, educational background and ability to work well with people.

Education • B. Tech. and M. S. Computer Science and Engineering

International Institute of Information Technology (IIIT), Hyderabad, India

– Master of Science (by Research) (Jul 2008 - Apr 2010)

Specialization : General Purpose Computation on the Graphics Processing Units

– Bachelor of Technology (2004 - 2008) CGPA 9.04

• Senior Secondary (2002) 88.5%

Shivaji Science College, Nagpur, Maharashtra

• Secondary (2002) 83.6%

Mary Immaculate Girls High School, Mumbai, Maharashtra

Experience • IBM Indian Research Lab, India (Jan 2011 - Present)

Working as a software research developer. Working on developing reverse time migra tion software on NVIDIA GPUs for oil and gas industry. Submitted research papers related to biological sequence alignment (GrABFAST), and graph algorithms (Ear Decomposition, Biconnected components) on GPU to top quality conferences.

• Cisco Systems, India (Aug 2010 - Jan 2011)

Worked as a software developer in Server Access Virtualization Business Unit.

• Research Assistant, CVIT, IIIT Hyderabad, India (Jun 2008 - Jun 2010)

Publications

• Sheetal Lahabar and P. J. Narayanan, Singular Value Decomposition on GPU us-

ing CUDA, 23rd IEEE International Parallel and Distributed Processing Symposium

(IPDPS 2009), Rome, Italy, May 2009.

• Sheetal Lahabar, Pinky Agrawal and P. J. Narayanan, Computer Vision on GPU, Na-

tional Academy of Sciences Journal (NAS 2008), Allahabad, India.

• Sheetal Lahabar, Pinky Agrawal and P. J. Narayanan, High Performance Pattern Recognition on GPU, 1st IEEE National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG 2008), Ahmedabad, India, Jan. 2008.

• Jyothish Soman, Ankur Narang and Sheetal Lahabar, Ear Decomposition and Bicon-

nected components on GPU using CUDA, Euro-Par, Rhodes Island, Greece, 2012. (Under Submission)

• Jyothish Soman, Ankur Narang and Sheetal Lahabar, GrABFAST: A GPU Acceler-

ated Fast Short Sequence Alignment Algorithm, International Conference on Parallel

Processing, Pittsburgh, PA, 2012. (Under Submission)

Dissertation Topic • General Purpose Computations on GPUs The rapid increase in the performance of graphics hardware have made GPU a strong candidate for performing many compute intensive tasks. They can be used to accelerate a wide range of scientic and engineering applications. My thesis primarily focuses on designing parallel algorithms on the GPU for compute intensive general purpose problems. We designed algorithms which exploits the SIMD GPU very well and give signicance performance.

• Singular Value Decomposition on GPU using CUDA: GPUs can be used for

solving linear algebra problems. We implemented the widely used Singular Value De-

composition(SVD) algorithm using the twin steps of Bidiagonalization followed by the

Diagonalization. SVD has many uses but very high algorithmic complexity. We de-

signed a hybrid algorithm which exploits the computational capabilities of GPU and

CPU. We achieved signicant speedup of upto 60 and 8 over the MATLAB and Intel

MKL implementation respectively.

• Ray tracing parametric patches on GPU: We utilize the GPUs for the purpose of

direct ray tracing of B‘zier patches which has the advantage of reduced memory usage

with increased precision. We are able to calculate exact hit points and perform per

pixel lighting for the rst time at interactive rates. We solve high degree polynomials in

parallel. We achieve 190 times speedup while solving the equations over CPU. The high

complexity of their implementation places more demands on the implementation.

Skillset

Graphics APIs Nvidia CUDA, OpenGL, Cg/GLSL

Programming Languages C, C++

Operating Platforms GNU/Linux, Windows

Web Languages HTML, CGI

Programming Environments GNU/GCC, Microsoft Visual Studio, Vim, Phoenix RDK

Libraries SDL, Qt, Glut, OpenCV, FANN, FFMPEG

Tools MySQL/MSSQL, Photoshop/Gimp

Research Projects

• High Performance Pattern Recognition on GPU: We utilize the GPUs for pattern

recognition problems specically the Parzen windows scheme for density estimation and

the Articial Neural Networks(ANN). We formulate the problems as matrix operations

which are inherently parallelizable. Our ANN implementation using CUDA achieved

170-210 times speedup over MATLAB and 230-270 times speedup over FANN library

on GTX 280.

• Computer Vision on the GPU: We proposed a design and implementation of a

general computer vision toolkit using the CUDA programming model. Our design is

based on OpenCV, a popular computer vision toolkit. We maintained similar program-

mer level representations. We implemented basic image processing techniques (ltering,

Canny edge detection), Optical Flow and Tracking and Background subtraction. It

provides a speedup of 10-30 on dierent operations on the Nvidia GTX 280 GPU.

• vGoogle: A google approach to video retrieval: We retrieve the key frames and

shots of a video containining a particular object or scene with the ease, speed and

accuracy with which Google retrieves text documents containing particular words. Pre-

processing allows Google to keep track of the ranked list of documents that corresponds

to every word which requires a merging of ranked documents for every query term

to form the result. We attempted to use this standard approach to recognition. We

were able to identify Visual vocabularies on dierent sets of videos. Technologies used:

MySQL, CGI, HTML.

Major Projects

• A smart phone book for mobile devices: We developed a smart phone book for

mobile devices which can store large volumes of data and also uses proactive browsing

techniques for quick searches in the phone book using Java and J2ME technology. Using

proactive browsing techniques, a mobile news reader is also developed which displays

the news articles based on certain keywords and links dierent news articles based on the

keywords. Technologies used: J2ME, Apache Tomcat.

• Image Layout Analysis and Reconstruction in HTML: Given an image and the

output HTML page details such as locations of text boxes, images, their attributes, the

text in the text boxes and their attributes, the project deals with writing a C/C++

routine to generate an HTML le which looks similar to the original image. Technologies

used: C++

Course Projects

• Computer Networks - File Transfer Protocol: We implemented the basic File

Transfer functions over the network. We used sockets for connecting with different

ports and for transferring data from the client to the server and vice versa.

• IT Workshop - Indic Spell Checker: We developed a Spell Checker GUI for Indian

languages using Aspell language dictionaries and developed an interface using Qt for

spell checking.

• MiddleWare Systems - Hospital Management System: We developed a database

system for a Hospital. The development process included problem specication, require-

ments identication, ER model diagram, Relational model design, populating a database

and Web-service design.

• Operating systems - Implementation of the BASH shell: Simulation of the

BASH shell of Linux using system calls, such as redirection operators, ls, etc.

• Database Management System - Sort Merge Join: We implemented the Sort

Merge Join to nd for each distinct value of the join attribute, the set of tuples in each

relation which displays that value.

• 2D Tetris Game and 3D simulations of a railway system: To write the tetris

game in 2D with functionalities of maintaining score and 3D train simulation on tracks

using OpenGL, Glut, SDL etc. Implemeted standard OpenGL library functions.

Courses Taken Computer Graphics, Digital Image Processing, Pattern Recognition, Computer Vision,

Signals and Systems, Theory of Computation, Cryptography, Data Compression, Linear Programming, Articial Intelligence, Introduction to Middleware Systems, Computer Organization, Spatial Informatics, Computer Networks, Compilers, Software Engineering, Principles of Programming Languages, Economics, Database Management Systems.

Achievements

• Have been in the Dean List for Academic Excellence in Monsoon 2004, Monsoon 2005,

Spring 2005, Monsoon 2006, Spring 2006 and Spring 2007

• Received the 23rd International Parallel and Distributed Processing Symposium 2009

student travel grant.

• Was awarded a scholarship by IARCS to attend IPDPS 2009 held in Rome, Italy.

Miscellaneous

• Attended NCVPRIPG 2008 at DAIICT Ahmedabad and IPDPS 2009 at Rome, Italy.

• Demonstrated the project Smart Phone Book for Mobile Devices and the Video Google

Project at the R&D at IIIT Hyderabad.

• Languages: English, Hindi, Marathi and German.

• Was one of the organizer of Cultural Events, at Felicity07, annual technical and cultural

festival at IIIT.

• Winner of inter house throw ball competitions and Western Group Dance.

Contact this candidate