Rahul Kumar

Contact Senior Undergraduate +91-971*******

Information Electrical Engineering Department acbn3q@r.postjobfree.com

Indian Institute of Technology (IIT) Delhi acbn3q@r.postjobfree.com

Research Machine Learning, Optimization, Computer Vision

Interest

Education Indian Institute of Technology (IIT) Delhi, India July 2010 - Present

Dual Degree : B.Tech in Electrical Engineering & M.Tech in Information and Communication Technology

CGPA : 8.59/10

Department Rank : 2/30

Scholastic

Achievements

Selected in the Top 30 of the country in the Indian National Astronomy Olympiad (INAO) and

attended the orientation camp for the International Olympiad on Astronomy & Astrophysics

(IOAA).

All India Rank 362 out of nearly 5,00,000 students in Joint Entrance Exam (IIT JEE).

All India Rank 134 out of nearly 1,000,000 students in All India Engineering Entrance

Examination (AIEEE).

All India Rank 15 in National Science Talent Search Examination (NSTSE).

Qualified National Standard Examination in Chemistry (NSEC) & got selected in Top 300

students of India in Chemistry Olympiad.

Qualified Regional Mathematics Olympiad (RMO) from Delhi Region and attended training

camp for Indian National Mathematics Olympiad (INMO).

Awarded for standing among Top 1% students of Delhi in Physics Olympiad by Indian

Association of Physics Teachers (IAPT).

Honours and

Awards

Awarded the Kishore Vaigyanik Protsahan Yojana (KVPY) fellowship for the period 2009-10.

KVPY is a prestigious award given to top 200 students of India by Indian Institute of Science

(IISc) Banglore to promote research.

Awarded the National Talent Search Examination Scholarship (NTSE). ). NTSE is a national

level scholarship organized by Government of India to identify and nurture talented students .

Awarded Merit Award for being among Top 7% of the department.

Awarded CBSE AIEEE Scholarship in All India Engineering Entrance Exam.

Awarded Scholarship for research purpose by Mahabir Pratap Singh Foundation for Education

and Research (MPSFER).

Participated in National Initiative on Undergraduate Science (NIUS) Physics Program at

HBCSE (TIFR) Mumbai.

Awarded as Vidalya Ratan (Gem of School) from school in class XII.

August 2012 – April 2013

Major Multi Label Incremental Random Forests

Projects Supervisor: Dr. Manik Varma, Microsoft Research India, (Visiting Researcher IIT-D)

Our project deals with the situation in which a huge amount of online data is coming & we need to kee p on

updating our random forest with new data. Training Random forest each time on whole set of old & new

data is computationally very expensive. We proposed a solution in which a new random forest will be

trained on new data only and combination of this new random forest and old random forest will be best

approximation to a random forest trained on whole set of data. Modified stochastic gradient boosting

technique was applied to learn the new random forest. Our technique gave almost equal accuracy saving

huge amount of training time. Some other function approximation techniques & update methods were also

explored.

May 2013 – July 2013

Automatic Cancer detection system

(Summer internship at Philips Healthcare Research, Bangalore)

Supervisor: Mr. Vipin Gupta, Philips Research India

Developed an automatic system to detect cervical cancer from colposcopy images. Our system detected

punctations (dot like structures) from the images of cervix which is a good indicator of cervical cancer.

Various components of my work:

Image processing techniques like Harris corner detection, morphological operations, filtering etc.

were implemented to get primary results of cancer detection.

CoHOG features from Gray Level Co-occurrence Matrix (GLCM) of cervix image were extracted

to predict cancer using a trained classifier.

Random forests were used as trained classifier to predict cance r from primary cancer detected

image. Further boosted random forests were used which enhanced the performance significantly.

Whole the system was implemented on .net framework using EmguCV (wrapper of OpenCV).

Parallel processing was implemented to make the system fast.

Sept 2013 – Nov 2013

Other Stock Market prediction using Newspaper Sentiments

Projects Supervisor: Prof. Parag Singla, IIT Delhi

I implemented a system for stock market prediction using newspaper sentiments. Machine learning

techniques like SVM, Naïve Bayes, GDA etc. were applied on various experimental setups with various

feature extraction techniques (i.e. bag of words, Chi -square, Mutual information etc.). Good Results were

observed as compared to traditional regression mo del.

Aug 2012 – Nov 2012

Face-recognition based Attendance System

Supervisor: Prof. Jayadeva, IIT Delhi

Worked on developing an image based attendance system for a large class based on face-detection and

recognition theory. I worked on enhancing the efficiency of existing methods for the given problem.

Techniques like Viola-Jones, PCA, SVM etc. were explored.

Aug 2013 – Present

Parametric Reshaping of Humans in a Video

Supervisor: Dr. Prem Kalra, IIT Delhi

My project involved changing the parameters like height, weight etc. of humans in a video. First human

body is segmented using Grab Cut & then a mesh is fitted to the human in the same pose. We use PCA to

change various parameters in the mesh & then motion retargeting is done to implement inverse kinematics

in the motion. Finally image warping (on foreground) & inpainting (on background) are done to produce

artifact free video.

Jan 2012 – April 2012

Indoor Navigation Solutions for the Visually Impaired

Supervisor: Prof. Ranjan Bose, IIT Delhi

Worked on an Indoor Navigation Technique for the Visually Impaired. Project involved tracking the

position of user and a java based application to create user interface. We designed a system which could

track the real time position of the user in the building, take the input from the user for where he wants to go

and give him voice commands accordingly.

Nov 2010 – Jan 2011

Line Following Robot Shooting at hidden destination

Designed an autonomous robot that could follow curved white lines, stop at locations with a wall on any

side of the track, and shoot TT balls on a lamp positioned in arena by measuring the range of lamp using its

light intensity. This was a design competition at Kshitij (Tech. Fest. of IIT Kharagpur) where we won the

'Best Fresher Team' award.

Relevant

Courses

Electrical Engineering: Signals and Systems, Digital Electronics, Analog Electronics, Control

Engineering, Communication Engineering, Electro-Magnetics, Digital Signal Processing, Digital

Communications, Information Theory

Mathematics: Real Analysis and Differential Equations, Linear Algebra and Matrix Theory,

Probability Theory and Stochastic Processes, Discrete mathematics

Computer Science: Data Structure and Algorithms, Computer Architecture, Computer Networks,

Operating systems, Numerical and Scientific Computing, Machine Learning, Neural Networks,

Digital Image Processing, Computer Vision

Skills

Programming Language: C, C++, C#, Python, Java, VHDL, HTML, JavaScript, ARM and 8086

Assembly Language.

Software Packages : Matlab, OpenCV, EmguCV, Xilinx ISE, Ngspice

Teaching and Teaching Physics: Worked as a part time teacher in a Coaching Institute and taught Physics to the students

Mentoring preparing for JEE(Joint Entrance Exam).

Avanti Mentor : Worked with Avanti – an NGO to help students with poor financial background. Visited

students in Delhi, taught them science & math and guided them for various aspects in life.

Student Mentor: Served as student mentor for freshman UG students of Electrical department.

Organizing Workshop : Organized workshop on Line following Robots and Astronomy & Astrophysics

for freshmen UG students.

Organizing Tryst : Worked as Event Coordinator for Electrical Engineering events in Tryst ( Technical

Fest of IIT Delhi) and organized four innovative competitions.

Technical Secretary: Worked as Technical Secretary of Electrical Engineering Society.

Hobbies Music, Athletics, Visiting Natural Beauties.

References Manik Varma: Researcher, Vision Graphics & Visualisation Group, Microsoft Research India,

+01-080-********, acbn3q@r.postjobfree.com

Parag Singla: Assistant Professor, Department of Computer Sc ience and Engineering, IIT Delhi,

acbn3q@r.postjobfree.com

Vipin Gupta: Sr. Scientist, Philips Research India Labs, acbn3q@r.postjobfree.com

Contact this candidate