Chhaya Methani

From Msplacements

E-mail: chhaya@research.iiit.ac.in, methani.chhaya@gmail.com
Contact No: +91 9866217451
Download Resume: Pdf_Format [1]

Career Objective

I am driven by an urge to take up challenging problems, analyze them and their various facets and subsequently develop innovative and efficient solutions. My objective is to work in the field of Computer Vision and Pattern Recognition and apply my academic experience to the latest research going on in the field.


Education

  • MS by Research (CSE), IIIT-Hyderabad, (expected December, 2009);
  • B. Tech. (CSE), IIIT- Hyderabad, June 2008, CGPA 8.73

Work Experience

  • Research Assistant, CVIT, May 2008 - till date.
    Project: Unconstrained Biometric Authentication for practical scenarios.
  • Teaching Assistant, Monsoon 2008 and Monsoon 2007.
    Pattern Recognition Systems.
  • Student Intern, CVIT, May - Jul 2006.
    Project: Hidden Markov Model based learning of human behavioural pattern.

Publications

[1] Chhaya Methani, Anoop Namboodiri “Pose Invariant Palmprint Recognition”, IEEE International Conference on Biometrics(ICB 09), Alghero, Italy, Jun, 2009

Academic Achievements

  • Was awarded a scholarship by IARCS to attend ICB '09 held in Alghero, Italy
  • Figured in the Dean's List for Academic Excellence in the following semesters: Monsoon 2004, Monsoon 2006, Spring 2006, Spring 2007 and Monsoon 2007.
  • Successfully completed an Elementary level course in German, organized by Goethe-Zentrum, a prestigious German Cultural Center in June 2008.
  • Was one among a small batch of students to be selected for attending Dr. A.P.J. Abdul Kalam's lecture series in IIIT-H


Conferences and Workshops attended

  • Attended and Presented my work in IEEE ICB '09, Alghero, Italy (Jun 2009).
  • Attended Indian Conference on Vision Graphics and Image Processing (ICVGIP 2008), held in Bhubneshwar, India (Dec. 2008)
  • TechVista, Microsoft Research India's annual research symposium, held in Chennai, India (Oct. 2008).
  • The first Indo-Israel Workshop on Computer Vision (WCV 2008), held in Hyderabad, India (Feb. 2008).
  • Workshop on Enabling Technologies and Applications for Launch Vehicles, held in Hyderabad, India (Oct. 2005).


Research Projects

Video Based Palmprint Recognition
(Advisor: Dr. Anoop Namboodiri)

    Video based biometric recognition is more suited for practical applications since videos are readily available and have more information as compared to a single image. But most of the videos available are low-quality and out of focus. This is a major issue for biometric applications especially for palmprint based recognition, as accuracy goes down considerably with deterioration of image quality. The major problem is the change in appearance caused due to the noisy nature of the images captured using low-end camera devices, but is compounded by specularity of the skin and blur due to motion and focus. In this paper, we propose a novel method to handle low-quality palmprint videos captured in an unconstrained setup. The method effectively combines the frames in the videos to reconstruct a single good image. Experiments conducted on palm videos taken from 100 users show a reduction in Equal Error Rate(EER) from 15.47% to 6% .

Pose Invariant Palmprint Recognition
(Advisor: Dr. Anoop Namboodiri)

    The aim of this project is to make a Palmprint Recognition System where the image capturing device is flexible. This means that the system should be able to handle cases when the subject's palm is not parallel to the imaging plane and has been shown at various different angles to the camera. This dramatically increases system's usability in real world conditions, as we can now work with a multi-purpose camera in uncontrolled circumstances, such as those mounted on a laptop, mobile device or those for surveillance. However, the performance of existing techniques for palmprint authentication falls considerably, when the camera is not aligned with the surface of the palm. The problems arise primarily due to variations in appearance introduced due to varying pose, but is compounded by specularity of the skin and blur due to motion and focus. By using an approach that uses Ransac Based Homography with the initial set of corresponding points coming from a combination of invariant feature points and a bag of potential interest points, we are able to reduce the recognition error from 22% to 8% on a dataset of 100 users.

Major Projects

  • Palmprint based Recognition System
    Faculty Advisor: Dr. Anoop Namboodiri
    (Honours Project, CVIT, IIIT-H.)
    The aim of the project was to authenticate individuals based on their palmprints. Because of the presence of strongly visible palm lines, palmprints can capture distinctive features even with low resolution images captured using web cameras saving memory & computation time. It is for this reason, that palmprint is fast emerging into a prominent and promising biometric modality. The subject is made to place his/her hand on a platform and the web camera used for image capture is fixed on top of the platform. Gabor Filter responses of the palm image are used as the feature vector. An accuracy of 99.2% was obtained on the standard PolyU University Dataset. A real-time running demo was also setup at the IIIT-H Research and Development Showcase (R&D Showcase) in 2008, and the system performed consistently well, authenticating 148 out of 150 test subjects correctly.
    Technologies used: C, Matlab
  • Video Google
    Faculty Advisor: Dr. C. V. Jawahar
    Semester Project for the course Computer Vision
    The aim of the project was to make a Google like search engine for videos using techniques of Content Based Image Retrieval (CBIR). Videos were indexed on the basis of their visual content ( viz. scenes, objects etc.), instead of the text labels given to images by users. The technique used is as follows: For each video in the database, a set of key frames are taken from the video; from each of these key frames, regions having invariant feature points are clustered to form something called visual words. We used SIFT as the feature descriptor. The visual words thus formed, are used to index the key frames and thus videos in the database. We worked on an in-house database containing News Videos having the meta data of date, stories etc. A GUI was also designed to assist the user with the query and to display the ranked results ordered as per the user's convenience.
    Technologies used: C, HTML, CSS, Java Applets.

  • Hidden Markov Model based learning of human behavioural pattern
    Faculty Advisor: Dr. Anoop Namboodiri
    Summer Intern Project, CVIT.
    Human actions can be broken down into basic categories like sitting, standing, bending, walking etc. Hidden Markov Models are probabilistic sequence models which can predict a future event based on the sequence seen so far. The basic actions of sitting, standing, walking etc. are made into nodes of the Hidden Markov Model and the actual bigger events like, jogging, sleeping etc can be predicted by the model. The classic paper on HMMs has been given by Rabiner. He also describes the three clasical problems related to HMM which were studied and implemented using C.
    Technologies used: C, Matlab

  • Effective Evaluation of Transfer Grammar Component
    Faculty Advisor: Dr. Rajeev Sangal and Dr. Dipti Misra
    Semester Project, LTRC, IIIT-H.
    In Natural Language Processing, Language translation is done by a combination of machine learning and a list of rules. The rules imbibe the grammatical structure of the sentences and hence guide the translation procedure. It is possible that the wrong rule is applied to a sentence or that 2 or more rules interfere with each other's working to give a wrong output. The system works to evaluate these discrepancies and iron them out.
    Technologies used: C, C++, QT, Java

  • Segmentation of Brain scan images and Retinal images for disease detection
    Faculty Advisor: Dr. Anoop Namboodiri
    Semester course project for Digital Image Processing.
    The project implements the well known Snakes Algorithm for segmenting CT scan brain images and retinal images. The algorithm works by taking initial points from the expert and then iteratively optimizing an energy function fitted on the initial points. This helps to detect the change in shape of the optical vein in an eye or the shape of the brain cerebral, so that the disease can be detected.
    Technologies used: MATLAB, MATLAB GUI Toolbox.

Other Projects

  • 2-D real-time implementation of the popular game TETRIS.
  • DES Encryption/Decryption algorithm.
  • Myftpserver and a prototype of the file system used in linux.
  • Middleware system to manage a hospital's database.

Studies

  • Artificial Neural Networks.
  • Rabiner's paper on Hidden Markov Models
  • Case Study In requirements Engineering
  • Study of various Encryption/Decryption algorithms
  • Various techniques used in Spatial Informatics.

Skillset

    Programming Languages  :- C, C++, Java, Matlab.
  • Operating Systems  : - GNU/Linux, Windows 95/98/2000/XP.
  • Scripting Languages  : - Perl, Shell.
  • Web Languages  : - HTML, CGI, PHP.
  • Libraries  : - Qt, OpenGL, GLUT , FANN, Socket Programming.
  • Publishing  : - LaTeXe

Advanced Courses Taken

Pattern Recognition, Computer Vision, Computer Graphics, Digital Image Processing, Theory of Computation, Cryptography and Network Security, Data Compression, Spatial Informatics, Linear Programming, Economics.

Academic Interests

Biometrics, Pattern Recognition, Computer Vision, Image Processing, Data Structures, Algorithms, Data Compression.

Leadership Roles/Positions of Responsibility

  • Finance Secretary: During 2007-08. Responsible for handling the {\bf Student Activity Fund(SAF)}. Requests for funds go through the Secretary, who is also responsible for managing and accounting for the expenses. Managed a budget of approx. 2 mn INRs.
  • Co-Founder and Coordinator of IIIT-H Dance Club: During 2006-07, I played an instrumental role in the inception of the Dance Club and working out a direction for the club and preparing an outline for the club activities. Also organized seminars and workshops, and conducted club meetings.
  • Part of Core Organizing Committee in Felicity: 2006-07. Conceptualized and organized the major night on-stage event Two-To-Tango, and organized the business plan presentation event called Bizwit. I was presented with a commendation certificate for my versatile role in Felicity '07.
  • Member of Apex Body: During 2007-08. Chosen to be a part of an Apex Decision Making Student body. Monitored fresher activities ensuring smooth transition, with the help of a group of student mentors.
  • Member of Student Parliament: During 2006-07 and 2007-08. The task was to act as a bridge between students and the administration and to effectively implement various institute policies.
  • Captain of Throwball Team: For 2 consecutive years, 2006-07 and 2007-08, I was the captain of Girls Throwball Team for my house, Agni. Led the team to victory in Inter House Throwball Tournament both the years starting a winning streak.

Extra Curriculur Achievements

  • Was awarded a commendation certificate for "outstanding contributions towards organizing events in cultural and sports arena" during my 4 year stay in my college, IIIT Hyderabad.
  • Received an appreciation certificate for organizing a month long Dance Workshop as part of IIIT-H Dance Club. I was also the co-founder and coordinator of the club.
  • Made it to IIIT-H Throwball A-Team in Inter College Tournament where we got second place. Was also a part of Table Tennis team in Inter College Tournament held in VIT, Vellore.
  • My team won Inter College Western Group Dance Competitions in Felicity 05, Felicity 06 and Felicity 07.
  • Won various Inter House Group Dance Competitions in both Western and Classical/Folk Dance categories.
  • Secured 10th position out of 1200 participants, in the stock market simulation game, Equity Pulse, organized at Threads, in IIIT in 2006.
Retrieved from "http://www.iiit.ac.in/msplacements/index.php/Chhaya_Methani"
Personal tools