Sreekanth Vempati

From Msplacements

Center for Visual Information Technology,
International Institute of Information Technology,
Gachibowli, Hyderabad-500032, INDIA.
E-mail: v_sreekanth@research.iiit.ac.in
HomePage: http://students.iiit.ac.in/~v_sreekanth
Contact No.:+91-9247 469 459

Career Objective
   I would like to work in the industry involving cutting edge Research and Development applying my learning ability, skills and experience in the fields of Computer Vision and Machine Learning.

Education

  • Master of Science by Research in Computer Science and Engineering, IIIT Hyderabad. (expected December, 2009)
    • Thesis: Semantic Concept Detection for large scale Image and Video Retrieval, under the guidance of Dr. C. V. Jawahar
  • B. Tech in CSE with Honors in Visual Information Processing
    • International Institute of Information Technology, Hyderabad. (CGPA 8.19/10), May 2008.
  • Senior Secondary
    • Gowtham Junior College, Gudavalli, Vijayawada. IPE (Percentage: 94.3), May 2004.
  • Secondary
    • St. Alphonsus High School, Nalgonda. SSC (Percentage: 88.2), April 2002.

Work Experience

  • Reseach Assistant in Center for Visual Information Technology (CVIT) during MS by Research (May, 2008 - till date).
  • Visiting Student, Visual Geometry Group, University of Oxford, UK, June - 2008.
  • System Administrator for Computational cluster for CVIT research lab, (July 2007 - till date).
  • Student Intern in CVIT, IIIT-H (May - Jul 2005 and 2006).
  • Student Intern in Digital Library of India(DLI), a project aiming at digitizing one million books (May - Aug 2006).

Academic Achievements

  • Opportunity to visit VGG, University of Oxford for doing a project in June 2008 with funding from UKIERI(UK India Education and Research Initiative).
  • Figured in the Dean's List for academic excellence in Monsoon 2007.
  • Secured All India rank 2393 out of nearly 4,00,000 applicants in AIEEE 2004.
  • Secured State rank 1201 out of nearly 1,00,000 applicants in EAMCET 2004(Engineering entrance test conducted in Andhra Pradesh).
  • Recipient of Pratibha Scholarship award from the Government. of Andhra Pradesh.

Major Publications

[1] Mihir Jain, Sreekanth Vempati, Pulla Chandrika. Example based video filters.
In Proceedings of the International Conference of Image and Video Retrieval(CIVR 2009), Santorini, Greece, July 2009. (PDF)

[2] James Philbin, Manuel Marin-Jimenez, Siddharth Srinivasan and Andrew Zisserman, Mihir Jain,
Sreekanth Vempati, Pramod Sankar and C. V. Jawahar. Oxford/IIIT TRECVID 2008 – Notebook paper.
In Proceedings of the TRECVID 2008 Workshop, Gaithersburg, USA, November 2008. (PDF)

Minor Publications

[3] Chetan Jakkoju, Sreekanth Vempati, B. Rakesh Babu, Vamshi Ambati and C. V. Jawahar. Book Reader Interface: Image Processing Issues.
In Proceedings of the 2nd International Conference on Universal Digital Libraries (ICUDL 06), Alexandria, Egypt, November 2006. (PDF)

Conferences/Workshops Attended

  • Indian Conference on Vision Graphics and Image Processing (ICVGIP) in 2008(held at Bhubaneshwar, India) and 2006(held at Madhurai, India).
  • TechVista, Microsoft Research India's annual research symposium, held in Chennai, India (Oct. 2008).
  • The First Indo-Israel Workshop on Computer Vision (WCV), held in Hyderabad, India (Feb. 2008).
  • Participated in Indian Robotics Challenge (IRoCha) conducted as a part of Internation Joint Conference on Artificial Intelligence(IJCAI) held in Hyderabad, India (January 2007).
  • Attended several local seminars (Barcamps, Startup saturday, etc) which discuss on the current web technologies.

MS Thesis

Semantic Concept Detection for large scale Image and Video Retrieval
(Advisor: Dr. C.V.Jawahar )
The broad objective of this work is to explore the state of the art computer vision and machine learning techniques for retrieval of videos containing semantic concepts like Mountain, People-dancing, Boats etc.,

Major Projects

  • Semantic Concept Retrieval in Videos
    ( Advisors: Dr. C.V.Jawahar, IIIT Hyderabad and Dr. Andrew Zisserman, University of Oxford, UK )
    The state-of-the-art techniques in feature extraction, classification methods and classifiers are integrated to provide a vision-only approach for semantic concept retrieval of videos. We have employed our algorithms in High level feature extraction task conducted as a part of TRECVID in the years 2008 and 2009. It consists of finding and ranking shots from the videos, having some objects such as boat, bus etc. and some human actions as well like, people eating, dancing etc. TRECVID has been becoming challenging ever since due to constant increase in training and testing data size (a total data of 400 hours~400GB in MPEG format). Cleaning of data, annotations, training of classifiers and testing on such large datasets is a real challenge. more (PDF)

  • Optimal Bag of Words
    ( Advisor: Dr. C. V. Jawahar )
    Recent research shows that Bag of words technique is one of the most successful approach for visual(image and video) search and classification tasks. In this work, we explore and see if we can go beyond these results with a set of optimizations in this tecnique.(Ongoing Work)

  • Example Based Video Filters
    ( Advisor: Dr. C.V.Jawahar )
    Many of the successful multimedia retrieval systems focus on developing efficient and effective video retrieval solutions with the help of appropriate index structures. In these systems, the query is an example video and the retrieved results are similar video clips which are available apriori in the database. In this work, we address a complementary problem of filtering a video stream based on a set of given examples. By filtering, we mean to detect, accept or reject the part of a video stream matching any of the given examples. This requires matching of example videos with the on-line video stream. Since the concepts of interest could be complex, we avoid explicit learning of a representation from the example videos to characterize the visual event present in the examples. We model the problem as simultaneous on-line spotting of multiple examples in a video stream. We employ a vocabulary trie for the filtering purpose and demonstrate the applicability of the technique in a variety of situations.more(PDF)

  • Overlaid Text Detection in broadcast news videos
    ( Advisor: Dr. C. V. Jawahar )
    The aim of this project was to detect text overlaid in the videos and use it for retrieval and annotation of the videos automatically without recognizing the text. For this purpose various text detection approaches were studied and implemented some of the algorithms. A text detection scheme was run on a large set of broadcast news video frames.

  • Video Google for broadcast news videos
    ( Advisor: Dr. C. V. Jawahar )
    (Semester Project for the course Computer Vision with a team of size 18)
    The project objective 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. SIFT is used for feature extraction. 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 designed for the user to give query and display the ranked results. more(PDF)

  • Face Recognition
    ( Advisor: Dr. C. V. Jawahar )
    Involves independent study of state-of-the-art Face recognition algorithms along with experiments on YALE database. We have used compared standard face recognition algorithms like Prinicpal Component Analysis and Linear Discriminant Analysis.

  • Building Tiff Plug-in for firefox and Imageserver for DLI
    ( Advisor: Dr. C. V. Jawahar )
    ( Summer Project with a team of size 3)
    The aim was to built tiff plug-in which enables firefox to view tiff images and image server to cater the requests.
    – TIFF image is the popular format for storage for document images due to less storage space. But these format is not supported by firefox. We here developed a plug-in which enables firefox to view tiff images. We have also developed Imageserver who's job is to serve other format of images ( PNG,etc ) from the stored tiff images on the fly. Currently whole project is deployed in Digital Library of India (DLI).more(PDF)

Other Projects

  • Enterprise Expert Search ( as a part of Information Extraction and Retrieval Course )
  • Implementation of Optimal Dimensionality Discriminant Analysis method and Its Application to Image Recognition ( as a part of Machine Learning Course )
  • Detection and removal of cracks from Digitized paintings done PDF( as a part of Digital Image Processing Course )
  • Developed IGL Library which aims at developing and implementing various GL functions ( as a part of Computer Graphics Course )
  • Built Tetris Game and Simulated Train ( as a part of Computer Graphics Course )
  • Simulation of Linear Hashing on databases ( as a part of Database Management Systems Course )
  • Simulated bash shell in Linux ( as a part of Operating Systems Course )

Studies

  • Large Scale Algorithms: Application of Graph based algorithms on large datasets(PDF) (as a part of Machine Learning Course)
  • Analysis on requirement engineering as a part of Software Engineering course
  • Various techniques used in Spatial Informatics.

Academic Interests

Machine Learning, Computer Vision, Image Processing, Information Extraction and Retrieval, Data Structures, Algorithms.

Skillset

  • Programming Languages  : - C, C++, Lex & Yacc
  • Operating Systems  : - GNU/Linux, Windows 95/98/2000/XP.
  • Scripting Languages  : - Python, Perl, Shell(BASH), Matlab.
  • Web Languages  : - HTML, XML, CGI, PHP.
  • Libraries  : - Qt, OpenGL, OpenCV.
  • Tools  : - MySQL/MSSQL, MS Office, Adobe Photoshop, Adobe Flash.
  • Publishing  : - LaTeX.
  • Text Editor  : - Vim.

Advanced Courses Taken

Machine Learning, Pattern Recognition, Computer Vision, Computer Graphics, Digital Image Processing, Artificial Neural Networks, Graph Theory, Database Management and systems, Information Extraction and Retrieval, Theory of Computation, Spatial Informatics, Operational Research, Linear Programming

Extra Curriculur

  • Worked as volunteer for Ashakiran, a social initiative by student of IIIT-H for teaching poor students.
  • Was part of organizing team for Felicity, 2007, IIIT-H's annual cultural and technical festival.
  • Volunteered for Excitement of Research in 2005 and 2006.
  • Presented my work in Research and Development showcase conducted at IIIT-H in 2009

Interests/Hobbies Photography, Linux, Web 2.0, Cricket, TT, Swimming, Travel, Music and Movies.

References

Retrieved from "http://www.iiit.ac.in/msplacements/index.php/Sreekanth_Vempati"
Personal tools