Naveen Kumar

From Msplacements

Home page: http://researchweb.iiit.ac.in/~naveenb/
Email: naveenb@research.iiit.ac.in
Contact No.: +91-9502-755-484




Education

  • Master of Science by Research in Computer Science and Engineering, IIIT Hyderabad. (July 2010)
  • B. Tech in CSE with Honors in Visual Information Processing
    • International Institute of Information Technology, Hyderabad. (CGPA 7.86/10), May 2008.
  • Senior Secondary
    • Gowtham Junior College, Gudavalli. IPE (Percentage: 96.0)
  • Secondary
    • Nehru Niketan School, Tenali. SSC (Percentage: 89.2)


Work Experience

  • Co-instructor of Computing Tools : Visualization, Spring 2008, IIIT-H
  • Teaching Assistant for Computer Graphics (CS3500), Spring 2008, IIIT-H.
  • Reseach Assistant at CVIT, IIIT-H (May 2008 - July 2010)
  • System Administrator for CVIT lab,(2006 - 2010).

Computer Skills

Graphics APIs : OpenGL, Cg/GLSL, CUDA
GUI APIs : Qt, SDL/GLUT
Programming Environments: Vim, Microsoft Visual Studio
Programming Languages : GNU/C, GNU/C++, Python
Scripting Languages: Bash, Matlab, Python, Action script 3.0
Documentation: LATEX, MS Office
Tools : MySQL, GIMP, Adobe Photoshop, Blender, Adobe Lightroom.
Web : HTML, PHP, CSS, Javascript.
Operating systems: GNU/Linux, Windows XP.

Publications

[1] Naveen Kumar Bolla, P J Narayanan “ Algebraic Splat Representation For Point Based Models”, IEEE Sixth
Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP 2008), 71 − 78, Bhubaneswar, India.

Projects

  • Algebraic Splats for High-Quality Rendering of Points
    Presented the methods to compute an algebraic-splat representation for a point set with a continuous level of detail scheme. The algebraic nature of the representation also enables efficient anti-aliasing of the models using an adaptive super-sampling for the silhouette pixels alone. We also present a rendering algorithm for algebraic splats that uses ray tracing of the basic primitives with per-pixel blending, adaptive anti-aliasing, and shadows.

  • Algebraic Splat Representation for Point-based models
    Proposed algebraic splats as a basic primitive of representation for point based models. Quadratic and cubic splats provide good quality and high rendering speed using far fewer primitives on a wide range of models. Our representation reduces the number of primitives needed by a factor 20 to 30 on most models and by a factor of over 100 on dense models like David with little or no drop in visual quality. Ray-casted the models like David at upwards of 200 frames per second on a commodity GPU

  • Ray-tracing of Metaballs
    Developed an algorithm for ray-tracing a million metaballs at interactive speeds using perspective girds and adaptive marching points algorithm. This has applications in rendering molecular data, water, fluid simulation.

  • Ray tracing implicit surfaces using CUDA
    General implicit surfaces can be ray traced efficiently on the GPU using Adaptive Marching Points algorithm. This problem offers us an ideal application for CUDA computation model, huge number crunching and virtually no data movement. We are able to extract 600-700 GFLOPS of performance from Nvidia GTX280.

  • A Proxy based compression of depth images
    Depthmaps differ from images qualitatively and new methods are needed to compress them. We used geometric proxy of the scene to compress multiple depthmaps of the scene. The residue depthmaps are compressed using various compression algorithms. A lot of experiments are conducted on the wide variety of the models with proxies at different LOD and camera configurations at various encoding and sampling of the depth maps.

  • vGoogle : A google approach to content based video retrieval
    Our method is based on work by Prof. Zisserman et. al. Major contributions are in developing methods/code for indexing and retrieval of visual words. A text retrieval like (inverted file systems and document rankings using tf-idf) approach is used for retrieving and ranking of results. Our system is efficient and quick (query time is less than few seconds for hours of news data).

  • Spam email classification using Support Vector Machines(SVM)
    Explored new approaches for spam email classification using machine learning techniques. We showed Support Vector Machines offer promising solution for learning and classification of spam.

  • IIIT’s Local Railway Information Site
    With an aim to provide faster and simpler access to train reservation information, a portal is developed to cache the reservation information. Information is automatically updated few times a day. This was developed using LAMP. In later years, lot of similar portals came up in Internet. (eg: erail).

  • Toolbox for Visual Interpretation
    A toolkit for viewing, adding, tagging and categorizing images from distrubted servers is developed using Qt. Basic image enhancement features are also provided.

  • Other Projects
    • A Semi supervised EM algorithm for Image retrieval as part of Machine Learning (CS5770). sample results (Monsoon 2007)
    • Range Aggregate Query problem involving Geometric operations as part of Computational Geometry (CS4175) code (Spring 2007)
    • Implementation for PDE based Image Inpainting as part of Digital Image Processing (CS4750) code report(Monsoon 2006)
    • OpenGl based Tetris game, 3D train simulation and openGL-like API as part of Computer Graphics (CS3500), Spring 2006.


Advanced Courses Taken

Computer Graphics,Machine Learning, Pattern Recognition, Computer Vision, Digital Image Processing, Computational Geometry, Database Management and systems, Information Extraction and Retrieval, Theory of Computation, Operations Research, Linear Programming

Academic Achievements

  • Recipient of prestigious Prathibha Scholarship from the State Government Andhra Pradesh during 2004-2008.
  • Is among the top 0.5% in All India Engineering Entrance Examination(AIEEE), 2004
  • Presented a paper at Indian Conference on Computer Vision, Graphics and Image Processing 2008
  • Recipient of Best student award of the year in X standard.
  • Conceptualized and mentored various projects for ITWS-III during my Masters.

Other

  • Designed and organized KodeKonundrums, on-line puzzle solving contest using programming/scripting skills as part of THREADS 2007.
  • Attended and Volunteered for Workshop on Computer Vision Graphics and Image Processing, 2006.
  • Attended and Volunteered for Asian Conference on Computer Vision (ACCV), 2006.

Other Interests/Hobbies

  • Photography, Linux, Tech Geek, Entrepreneurship, Web 2.0, Cricket,Drawing, Music and Movies.
Retrieved from "http://www.iiit.ac.in/msplacements/index.php/Naveen_Kumar"
Personal tools