CS6350: Computer Vision

January - May, 2018

Course Contents

  • Objectives:
    • Computer Vision focuses on development of algorithms and techniques to analyze and interpret the visible world around us. This requires understanding of the fundamental concepts related to multi-dimensional signal processing, feature extraction, pattern analysis visual geometric modeling, stochastic optimization etc. Knowledge of these concepts is necessary in this field, to explore and contribute to research and further developments in the field of computer vision. Applications range from Biometrics, Medical diagnosis, document processing, mining of visual content, to surveillance, advanced rendering etc.
  • References

    Textbooks
    Richard Szeliski, Computer Vision: Algorithms and Applications, Springer-Verlag London Limited 2011.

    Computer Vision: A Modern Approach, D. A. Forsyth, J. Ponce, Pearson Education, 2003.


    References
    Richard Hartley and Andrew Zisserman, Multiple View Geometry in Computer Vision, Second Edition, Cambridge University Press, March 2004.
    K. Fukunaga; Introduction to Statistical Pattern Recognition, Second Edition, Academic Press, Morgan Kaufmann, 1990.
    R.C. Gonzalez and R.E. Woods, Digital Image Processing, Addison- Wesley, 1992.

    Journals
    IEEE-T-PAMI (IEEE Transactions on Pattern Analysis and Machine Intelligence).
    IJCV (International Journal of Computer Vision) - Springer.

    Lecture Slides

    Topic-Wise References

    1 Introduction  Download Slide
    2 Neighborhood and Connectivity of pixels Download Slide
    3 3D transformations and projection Download Slide
    4 Fourier Theory and Filtering in spatial and spectral domains

    Download Slide

    5 Enhancement Download Slide
    6 Histogram based image processing Download Slide
    7 Concepts in Edge Detection Download Slide
    8 Hough Transform Download Slide
    9 Image segmentation Download Slide
    10 Texture analysis using Gabor filters Download Slide
    11 SCALE-SPACE - Theory and Applications Download Slide
    12 Local Feature Detectors and Descriptors Download Slide
    13 Motion Download Slide
    14 Motion Detection and Tracking Download Slide
    15 Shape from Shading Download Slide
    16 Pattern Recognition Download Slide
    17 Wavelet transform Download Slide
    18 Morphology Download Slide

    Download

    Tracking Videos Download Schedule

    Term Project Assignment



     

    Term Project Assignment List
    SL No. Problem Statement Performance based marks (indicative) Group IDs Alloted (by preference, FCFS)
    Details
    Preference No.
    Satisfactory Good Excellent I II III
    1 Enhancing Super-resolved Images, from those with Low-resolution and Low-contrast 18 28 36 -- -- -- Download Slide
    2 Saliency detection from images of multiple objects: Top-down/Bottom-up or combination 15 30 38/40 -- -- -- Download Slide
    3 Comparative Study of the Performances of very recent Feature Extractors, used for Detection, Matching and Recognition (Image or Video) 20 34 41 -- -- -- Download Slide
    4 3-D jigsaw puzzle solving for structure reconstruction from assorted parts 20 34 40 -- -- -- Download Slide
    5 Semi-supervised clustering for organizing large image/video datasets 20 32 41 -- -- -- Download Slide
    6 Domain Adaptation using Kernel (or Manifold) Alignment for Object Categorization on ImageNet 20 32 41 -- -- -- Download Slide
    7 Face Recognition from LFW/YouTube/Multi-PIE faces, in presence of occlusion and pose 20 36 41 -- -- -- Download Slide
    8 Deep Learning on ImageNet for Object Recognition 20 35 42 -- -- -- Download Slide
    9 Online CBIR tool to work indoor for a large category of handheld and personal items 20 32 40 -- -- -- Download Slide
    10 Video Event Categorization (UCF-101, HMDB-51) 20 34 41 -- -- -- Download Slide
    11 Video/Image Captioning (MS-COCO, Flickr30k) 20 35 42 -- -- -- Download Slide
    12 Discriminatory features (hand-crafted or deep learning based) using subspace manifolds or equivalent 25 35 42 -- -- -- Download Slide
    13 Panorama Generation from Videos with Significant Blur 25 35 42 -- -- -- Download Slide
    14 Retail Product Recognition on Supermarket Shelves 25 35 42 -- -- -- Download Slide

    Tutorial

    Tutorial No. Date Time
    1 25/01/2018  10:00-10:50    
    2 12/02/2018 12:00-12:50  
    3 14/03/2018 16:50-17:40  
    4 04/04/2018 17:30-18:30  
    5 20/04/2018 09:00-09:50  

    Schedule

    Academic Calender Download Schedule

    Marks Distribution

    Logistic Details


    Important Dates (Tentative)
    Finalizing the groups by students

    31/01/2018
    Choice of TPA by students
    10/02/2018
    TPA Assignment to each group

    14/02/2018
    TA interaction for TPA

    20/03/2018
    Interim TPA Review(optional)

    29/03/2018 17:00-18:00
    Final TPA Demo

    27/04/2018 , 14/05/2018
    Tutorial Dates

    25/01/2018, 12/02/2018, 14/03/2018, 04/04/2018 and 20/04/2018
    Extra Classes (VENUE: BSB 361)

    31/03/2018 1100-1230 HRS, 14/04/2018 1100-1230 HRS
    Mid Semester Exam 09/03/2018 09:00-10:00  
    End Semester Exam 09/05/2018 09:00-12:00  


    Announcements

    Tutorial-5 solutions - Download May 06, 2018
    Tutorial-4 solutions - Download May 06, 2018
    Tutorial-3 solutions - Download May 06, 2018
    Slides 1-13 of "Concepts in Edge Detection" are part of the self-study topics Mar 15, 2018
    Tutorial-3 is scheduled on Mar 14, 2018 (Wed 1645-1745 HRS). The syllabus only includes topics from "3D Transformation and Projection" slides taught in the class and any self-study topics from this area. Mar 10, 2018
    Tutorial-2 solutions - Download Mar 07, 2018
    TPA allocation - Download Feb 21, 2018
    Tutorial-2 is scheduled on Feb 12, 2018 (Mon 1200-1250 HRS). Bring your tutorial notebook.
    Portions include Fourier Theory, Convolutions, Histogram based image processing, 3D transformations and projection (first 27 slides).
    Feb 10, 2018
    Google form for filling TPA choices - Link
    Last date for filling - Feb 15, 2018
    Feb 10, 2018
    Tutorial-1 solutions - Download Feb 9, 2018
    Contrast-Limited Adaptive Histogram Equalization (CLAHE) and Weiner filter are part of self-study. Feb 5, 2018
    Minor modifications/updates of slides may happen from time to time. Feb 4, 2018
    Self-study topics (listed on the course webpage and announced in the class) and pre-requisites are part of the syllabus for the tutorials and exams. Feb 2, 2018
    Please go through the first 22 slides of "Histogram-based Image Processing". You are supposed to come prepared for today's (Jan 31, 2018) class. Jan 31, 2018

    Self-study


    Self Study Topics

    Pre mid-sem
    1 Neighborhood and Connectivity of pixels Download Slide
    2 First part :
    Fourier Theory and Filtering in spatial and spectral domains

    Download Slide

    3 First 22 slides :
    Histogram based image processing

    Download Slide

    Post mid-sem
    4 Slides 44-59 :
    Pattern Recognition

    Download Slide

    5 First 25 slides :
    Local Feature Detectors and Descriptors

    Download Slide