Pratyush Bhatt

From Msplacements

E-mail: bhatt@research.iiit.ac.in
Contact No: +91 9866212689
Download Resume: Pdf_Format [1]

Career Objective

To apply my academic experiences and interpersonal skills to build a research career in the field of computer vision and related areas.

Education

• MS by Research (CSE), IIIT-Hyderabad, (expected December, 2009);
• B. Tech. (CSE), IIIT- Hyderabad, June 2008, CGPA 9.02

Work Experience

• Teaching Assistant for C Programming in Monsoon 2006
• Teaching Assistant for Pattern Recognition in Monsoon 2008
• Teaching Assistant for Advanced Problem Solving in Monsoon 2009
• Worked as a Research Assistant during my Masters, from May 2008 onwards, at the Center for Visual Information Technology, IIIT-H

Publications

[1] Pratyush Bhatt, Anoop Namboodiri “Hierarchical Local Maps for Robust Approximate Nearest Neighbour Computation”, 7th International Conference on Advances in Pattern Recognition (ICAPR 2009),Feb. 4-6, 2009, Kolkotta, India.

Conferences and Workshops

• Attended and Presented my work in ICAPR '09, Kolkata, India (Feb 2009).
• Attended Indian Conference on Vision Graphics and Image Processing (ICVGIP 2008), held in Bhubneshwar, India (Dec. 2008)

Research Projects

Hierarchical Local Maps for Robust Approximate Nearest Neighbor Computation
Advisor: Dr. Anoop Namboodiri

We propose a novel method for fast nearest neighbors retrieval in non-Euclidean and non-metric spaces. We organize the data into a hierarchical fashion that preserves the local similarity structure. A method to find the approximate nearest neighbor of a query is proposed, that drastically reduces the total number of explicit distance measures that need to be computed. The representation overcomes the restrictive assumptions in traditional manifold mappings, while enabling fast nearest neighbor’s search. Experimental results on the Unipen and CASIA Iris datasets clearly demonstrates the advantages of the approach and improvements over state of the art algorithms. The algorithm can work in batch mode as well as in sequential mode and is highly scalable.

Major Projects

• Color Based Joint Detection
  Faculty Advisor: Dr. Anoop Namboodiri
  Project for Phillips Research, India
  Developed a primary level system for Phillips Research India, which recognizes human body pose with simple motion based tactics. Color markers were used to capture joints in body and stick model was fitted to simulate the human action
  Technologies used: C, Matlab


• Fingerprints Enhancement and Matching
  Faculty Advisor: Dr. Anoop M. Namboodiri
  Honours Project - I
  This project involved the building of a fingerprint enhancement and matching system. Given any degraded fingerprint image, it enhances the image by segmenting the ridges and then estimating the local orientation of ridges and finally enhancing it using oriented filters. For matching the minutia of the enhanced image are extracted and using registration and correlation techniques find the best match with the fingerprints already stored in the database. This project employed intensive image processing and minutia extraction and matching techniques.
  Technologies used: C, Qt


• Sequential MCMC for Bayesian Model Selection
  Machine Learning Course Project
  The project required the thorough understanding of Sequential Monte-Carlo Algorithm and its application to parameter prediction and estimation in a time varying model. MCMC itself have several applications which require sampling from posterior distribution like body pose estimation.
  Technologies used: Matlab


• Printing Inspection System
  Faculty Advisor: Dr. Anoop M. Namboodiri
  Project for Printings Den, Hyderabad, India.
  An industrial project given to us by Printings Den which involved the building of an automated tool that would detect any color or textual error in printing of labels of medicines and liquor bottles. The image of the label to be checked was taken through a high resolution camera, registered with the standard sample and then matched with the standard samples to check for any errors or discrepancies.
  Technologies used: C, Matlab.



Other Projects

• 3D Train modeling in openGL in 4th semester Graphics Course.
• Implemented Optical Flow on GPU using CUDA.
• Robosoccer Toolkit For Offline Game Analysis.
• Instant Messenger using Socket programming.

Skillset

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

Advanced Courses Taken

Machine Learning, Pattern Recognition, Computer Vision, Computer Graphics, Digital Image Processing, Information Extraction and Retrieval, Spatial Informatics, Advanced Linear Programming.

Academic Interests

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

Academic Achievements

• Figured in IIIT Deans List for Academic Excellence for all semesters.
• Project on Instant Messenger was kept in Excitement of Research (ExOR) 2006.
• Project on Robosoccer was kept in Research and Development(R&D) showcase 2006
• Secured 4th Rank in All India Inter DPS Maths Olympiad.
• Received Gold medal from school for appearing in Merit List for 9 consecutive years.

Extra Curricular

• Midfielder of IIIT-H Football Team
• Secured third rank in Inter-house Chess Competition
• 1st Runner up in Inter-house Badminton Competition
• Secured 2nd position from India, in a smart coding contest, Time Limit Exceeded, organized at Threads, in IIIT in 2008.