PROGRAM M. Tech in · 2020. 3. 11. · PROGRAM M. Tech in Computer Vision and Image Processing ... (M. Tech.) in Computer Science and Engineering, Artificial Intelligence and Data

1

PROGRAM

M. Tech in

Computer Visionand ImageProcessing

Department of Computer Science and Engineering

CURRICULUM AND SYLLABUS(For 2016 Admissions)

Table of Contents

Program Outcomes (POs)...................................................................................................5

Curriculum..........................................................................................................................6

Evaluation Pattern ............................................................................................................11

Syllabi...............................................................................................................................14

2

Vision of the Institute

To be a global leader in the delivery of engineering education, transforming individuals to become

creative, innovative, and socially responsible contributors in their professions.

Mission of the Institute:

* To provide best-in-class infrastructure and resources to achieve excellence in technical

education,

* To promote knowledge development in thematic research areas that have a positive impact on

society, both nationally and globally,

* To design and maintain the highest quality education through active engagement with all

stakeholders –students, faculty, industry, alumni and reputed academic institutions,

* To contribute to the quality enhancement of the local and global education ecosystem,

* To promote a culture of collaboration that allows creativity, innovation, and entrepreneurship to

flourish, and

* To practice and promote high standards of professional ethics, transparency, and

accountability.

Vision of the Department

To be acclaimed internationally for excellence in teaching and research in Computer Science & Engineering, and

in fostering a culture of creativity and innovation to responsibly harness state-of-the-art technologies for societal

needs.

3

Mission of the Department

1. To assist students in developing a strong foundation in Computer Science and Engineering by providing

analytical, computational thinking and problem solving skills.

2. To inculcate entrepreneurial skills to develop solutions and products for interdisciplinary problems by

cultivating curiosity, team spirit and spirit of innovation.

3. To provide opportunities for students to acquire knowledge of state-of-the-art in Computer Science and

Engineering through industry internships, collaborative projects, and global exchange programmes with

Institutions of international repute.

4. To develop life-long learning, ethics, moral values and spirit of service so as to contribute to the society

through technology.

5. To be a premiere research-intensive department by providing a stimulating environment for knowledge

discovery and creation.

The Department of Computer Science and Engineering was established on 7th October 1996 with seven faculty

members . In nearly two decades, it has grown into one of the major departments in the Amrita Vishwa

Vidyapeetham, with a dedicated team of 70+ experienced and qualified faculty members demonstrating

excellence in teaching and research. Currently the department offers Bachelor of Technology (B. Tech.) in

Computer Science and Engineering, Master of Technology (M. Tech.) in Computer Science and Engineering,

Artificial Intelligence and Data science. The department also offers Ph.D. programmes in thrust areas. The

department has attracted the best engineering aspirants and research scholars across the country.

As the nature of Computer Science and Engineering has a never-ending cycle of innovation rooted in it, our focus

is on increasing the standard of the teaching and learning to gain international importance in research and

academics. Looking at the global perspective the department has identified a few thrust areas for research and

development: Computational Intelligence and Information Systems, Artificial Intelligence, Data Science,

Networks and Internet of Things (IoT).

The department has set up a research hub with state-of-the-art facilities at Amrita Multi Dimensional Data

Analytics Lab, Amrita Cognizant Innovation Lab, Signal Processing and Mobile Applications Lab, Networks and

Internet of Things (IoT) Lab and Computational thinking for Research and Education (CORE) Lab, Smart space

lab. Cisco India Pvt Ltd has sponsored ThingQbator, a lab where focused research in IoT takes place. Only five

Universities in India have this facility.

4

This has enabled a large number of faculty getting involved in research and producing high quality solutions.

Program Outcomes (POs)

PO1- An ability to independently carry out research /investigation and development work to solve practicalproblems

PO2: An ability to write and present a substantial technical report/document

PO3- Students should be able to demonstrate a degree of mastery over the area as per the specialization of theprogram. The mastery should be at a level higher than the requirements in the appropriate bachelor program

5

6

M. TECH in COMPUTER VISION AND IMAGE PROCESSING


CURRICULUM(For 2016 Admissions)

7

First Semester

CourseCode

Type Course L T P Cr

16MA605 FCLinear Algebra and Partial DifferentialEquations 4 0 0 4

16CV601 FCMultidimensional Data Structures andAlgorithms 3 0 1 4

16CV603 FC Multidimensional Digital Signal Processing 3 0 0 3

16CV605 SC Digital Image Processing 3 0 1 4

16CV607 SC Digital Video Processing 3 0 1 4

16CV609 SCDevelopment Tools for Image andVideo Processing 0 0 1 1

16HU601 HU Cultural Education*P/F

Credits 20

* Non Credit CourseSecond Semester

CourseCode


16MA606 FC Random Processes and Optimization 4 0 0 4

16CV602 SC Image Analysis 3 0 1 4

16CV604 SC Pattern Recognition and Machine Learning 3 0 1 4

16CV606 SC Computer Vision 3 0 1 4

16CS625 SC Research Methodologies and Seminar 1 0 1 2

16EN600 HU Technical Writing* P/F

Credits 18* Non Credit Course

Third Semester

CourseCode


E Elective I 3 0 0 3

E Elective II 3 0 0 3

16CV798 P Dissertation 10

Credits 16

Fourth Semester

CourseCode


16CV799 P Dissertation 12

Credits 12

Total Credits 66

List of Courses

Foundation Core

CourseCode

Course L T P Cr

16MA605 Linear Algebra and Partial Differential Equations 4 0 0 4

16CV601 Multidimensional Data Structures and Algorithms 3 0 1 4

16CV603 Multidimensional Digital Signal Processing 3 0 0 3

16MA606 Random Processes and Optimization 4 0 0 4

8

Subject Core

CourseCode

Course L T P Cr

16CV605 Digital Image Processing 3 0 1 4

16CV607 Digital Video Processing 3 0 1 4

16CV609 Development Tools for Image and Video Processing 0 0 1 1

16CV602 Image Analysis 3 0 1 4

16CV604 Pattern Recognition and Machine Learning 3 0 1 4

16CV606 Computer Vision 3 0 1 4

16CS625 Research Methodologies and Seminar 1 0 1 2

Electives

Course

CodeCourse L T P Cr

16CV701 Modeling and Visualization 3 0 0 3

16CV702 Video Analytics 3 0 0 3

16CV703 Advanced Computer Vision 3 0 0 3

16CV704 Machine Vision 3 0 0 3

16CV705 Medical Image Analysis 3 0 0 3

16CV706 Content Based Image and Video Retrieval 3 0 0 3

16CV707 Document Image Analysis 3 0 0 3

16CV708 Computational Intelligence for Image Processing 3 0 0 3

16CV709 Image Fusion 3 0 0 3

9

16CV710 Multimedia Security 3 0 0 3

16CV711 Visual Pattern Analysis and Modeling 3 0 0 3

16CV712 GPU Architecture and Programming 3 0 0 3

Project Work

CourseCode

Course L T P Cr

16CV798 Dissertation 10

16CV799 Dissertation 12

10

11

Evaluation Pattern

50:50 (Internal: External) (All Theory Courses)

Assessment Internal External

Periodical 1 (P1) 15

Periodical 2 (P2) 15

*Continuous Assessment (CA) 20

End Semester 50

80:20 (Internal: External) (Lab courses and Lab based Courses having 1 Theory hour)


*Continuous Assessment (CA) 80

End Semester 20

70:30(Internal: External) (Lab based courses having 2 Theory hours/ Theory and Tutorial)

Theory- 60 Marks; Lab- 40 Marks


Periodical 1 10

Periodical 2 10

*Continuous Assessment

(Theory) (CAT)

10

Continuous Assessment (Lab)

(CAL)

40

End Semester 30

65:35 (Internal: External) (Lab based courses having 3 Theory hours/ Theory and Tutorial)

*CA – Can be Quizzes, Assignment, Projects,

and Reports.

12

Theory- 70 Marks; Lab- 30 Marks


Periodical 1 10

Periodical 2 10

*Continuous Assessment

(Theory) (CAT)

15

Continuous Assessment (Lab)

(CAL)

30

End Semester 35

Letter

GradeGrade Point Grade Description

O 10.00 Outstanding

A+ 9.50 Excellent

A 9.00 Very Good

B+ 8.00 Good

B 7.00 Above Average

C 6.00 Average

P 5.00 Pass

F 0.00 Fail

Grades O to P indicate successful completion of the course

( C xGr )i iCGPA =

C i

Where

Ci = Credit for the ith course in any semester

Gri= Grade point for the ith course

13

Cr. = Credits for the Course

Gr. = Grade Obtained

14

M. TECH in COMPUTER VISION ANDIMAGE PROCESSING


SYLLABI(For 2016 Admissions)

16MA605 LINEAR ALGEBRA AND PARTIAL DIFFERENTIAL EQUATIONS 4-0-0-4

Vector Spaces: Vector spaces - Sub spaces - Linear independence - Basis - Dimension - Inner products -Orthogonality - Orthogonal basis - Gram Schmidt Process - Change of basis - Orthogonal complements -Projection on subspace - Least Square Principle.

Linear Transformations: Positive definite matrices - Matrix norm and condition number - QR-Decomposition - Linear transformation - Relation between matrices and linear transformations - Kerneland range of a linear transformation - Change of basis - Nilpotent transformations - Similarity of lineartransformations - Diagonalisation and its applications - Jordan form and rational canonical form.Introduction to Normed Linear space, Banach spaces and Hilbert space.

Partial Differential Equations: Basic definitions, Model Equations: Elliptic, Parabolic and Hyperbolic

PDEs, Solving PDEs Numerically- Elliptic, Parabolic and Hyperbolic Equations. Finite Element Method.

COURSE OUTCOMES

16MAT608-CO1 Understand and apply vector space concepts

16MAT608-CO2 Understand and apply Linear Transformation

16MAT608-CO3 Apply Diagonalisation for different types of spaces

16MAT608-CO4 Understand Model equations for elliptic parabolic and hyperbolic planes.

TEXT BOOKS / REFERENCES:

* Howard Anton and Chris Rorres, “Elementary Linear Algebra”, Tenth Edition, John Wiley andSons, 2010.

* Gilbert Strang, “Linear Algebra and Its Applications”, Fourth Edition, Cengage, 2006.* Justin Solomon, “Mathematical Methods for Computer Vision, Robotics, and Graphics”,

Stanford University, 2013.* Lawrence C. Evans, “Partial Differential Equations”, American Mathematical Society, 2010.

* Gilles Aubert, Pierre Kornprobst, “Mathematical Problems in Image Processing: PartialDifferential Equations”, Springer, 2006.

16CV601 MULTIDIMENSIONAL DATA STRUCTURES AND ALGORITHMS 3-0-1-4

Asymptotic Notation - Standard notations and common functions - Recurrences – Substitution and Master

Method - Amortized Analysis – Aggregate Method

Introduction to Stacks and Queues, Linked list, B-Tree, B+ Tree. Multidimensional Data: Introduction –

Need for Multidimensional Data Structures. Multidimensional Point Data Structures: Point Quadtrees –

Trie-based Quadtrees – KD Trees (Point and Tries). Object based and Image based Representations:

1 5

Interior bases representations: Cells and tiling, Ordering Space

– Blocks - Non Orthogonal Blocks - Region Quadtrees - Region Octrees. Boundary Based

Representations: Boundary Model - Voronoi Diagrams - Properties - Delaunay Graphs - Delaunay

Triangulations.

COURSE OUTCOMES

16CV601-CO1 Understand and apply vector space concepts

16CV601-CO2 Understand and apply stack,Queue,Linked list

16CV601-CO3 Understand Multidimensional data structures and Multidimensional point structures

16CV601-CO4 Understand and apply Object based representations and point based representations

16CV601-CO5 Understand Boundary based representations for image processing

TEXT BOOKS/ REFERENCES:

1. Hanan Samet, “Foundations of Multidimensional and Metric Data Structures”, Morgan

Kaufmann, San Fransisco, 2006.

2. Michael T Goodrich and Roberto Tamassia, “Algorithm Design: Foundations, Analysis and

Internet Examples”, John Wiley and Sons, 2001.

3. Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest and Clifford Stein, “Introduction

to Algorithms”, Second Edition, The MIT Press, 2001.

16CV603 MULTIDIMENSIONAL DIGITAL SIGNAL PROCESSING 3-0-0-3

One Dimensional and Two Dimensional Signals and Systems- Separable Signals- Periodic Signals -

General Periodicity - 1D & 2-D Discrete Space Systems- 1 D & 2D Convolution. Continuous Space

Fourier Transform - Sampling in One and Two Dimensions - Ideal Rectangular Sampling – Sampling

Theorem - General Case – Change of Sampling Rate - Sampling Lattice – Reconstruction – Down

Sampling and Up sampling by integers.

Discrete Space Transforms – 1D & 2D Discrete Fourier Series - 1D & 2-D Discrete Fourier transform-

Properties – Discrete Time Fourier Transform- Short Time Fourier Transform – Fast Fourier Transform -

Wavelet Transform - Gabor Transform.

Two Dimensional Systems and Z-Transforms - 2D Spatial Systems – Z-Transforms - Regions of

Convergence - Linear Mapping of Variables - Inverse Z-Transform.

Filter Design Fundamentals - Ideal and Finite order Filters - Two Dimensional Filter Design - FIR and IIR

filter design - Window Functions - Rectangular and Rotated windows.

COURSE OUTCOMES

1 6

16CV603-CO1 Understand the properties of signals in spatial domain and apply them to appropriatescenario

16CV603-CO2 Apply basic operations on the signals and understand their constraints

16CV603-CO3 Understand the properties of signals in frequency domain and apply them to appropriatescenario

16CV603-CO4 Analyze the 1D and 2D signals in transformed domain

16CV603-CO5 Design appropriate filters and apply them to signals

16CV603-CO6 Evaluate the signal processing algorithms using modern tools

TEXT BOOKS/ REFERENCES:

1. Dan E. Dudgeon and Russell M. Mersereau, "Multidimensional Digital Signal Processing",

Prentice-hall, 1984.

2. John W. Woods, “Multidimensional Signal, Image, and Video Processing and

Coding”, Second Edition, Academic Press, Elsevier Inc. 2012.

3. Steven W. Smith, “Digital Signal Processing – A Practical Guide for Engineers and Scientists”,

Newnes Elsevier Science, 2013.

4. Alan V Oppenheim and Schafer Ronald W, “Digital Signal Processing”, Prentice Hall, 2008.

5. Sanjit K Mitra, “Digital Signal Processing: A Computer-Based Approach”, Third Edition,

McGraw-Hill Companies, 2005.

16CV605 DIGITAL IMAGE PROCESSING 3-0-1-4

Image Representation and Properties: Introduction - Image Representation - Image Digitization - Digital

Image Properties – Discrete Fourier Transform - Image Pre-Processing in Spatial and Frequency Domain:

Pixel Brightness Transformation - Geometric Transformations - Local Preprocessing - Image Smoothing

– Edge Detectors - Corner Detectors - Image Restoration.

Image Segmentation: Thresholding – Edge- Based Segmentation – Region Based Segmentation, Mean

shift segmentation, Graph cut algorithm– Matching – Evaluation Issues in Segmentation, Watersheds.

Color Image Processing: Color Fundamentals – Color Models – Pseudocolor Image Processing – Basics

of Full Color Image Processing – Color Transformations – Smoothing and Sharpening – Color

Segmentation – Noise in Color Images.

COURSE OUTCOMES

16CV605-CO1 Understand fundamental principles of signal and image processing

16CV605-CO2 Understand and apply spatial domain transformation for image enhancement

16CV605-CO3 Understand Frequency domain processing and apply it for image enhancement

16CV605-CO4 Understand the fundamental image segmentation techniques for feature extraction

16CV605-CO5 Understand the use of various feature descriptors for image description

1 7


1. Rafael C. Gonzalez and Richard E. Woods, “Digital Image Processing”, Third Edition,

Pearson Education, 2009.

2. Milan Sonka, Vaclav Hlavac and Roger Boyle, “Image Processing, Analysis and Machine

Vision”, Third Edition, Cengage Learning, 2007.

3. William K. Pratt, “Digital Image Processing”, Fourth Edition, Wiley Interscience, 2007.

4. Anil K Jain, “Fundamentals of Digital Image Processing”, Prentice Hall, 1989.

16CV607 DIGITAL VIDEO PROCESSING 3-0-1-4

Video Formation, Perception and Representation: Color Perception and Specification-Video Capture and

Display-Analog Video Raster-Analog Color Television Systems-Digital Video. Time Varying Image

Formation Models: Three Dimensional Motion models, Geometric Image Formation, Photometric Image

Formation.

Two Dimensional Motion Estimation: Optical Flow-General Methodologies- Pixel Based Motion

Estimation-Block Matching Algorithm-Deformable Block Matching Algorithms-Mesh Based Motion

Estimation-Global Motion Estimation-Region Based Motion Estimation-Multiresolution Motion

Estimation.

Video compression standards- Standardization-Video Telephony with H.261 and H.263,MPEG-1,MPEG-

2,MPEG-4/H.264, Compressed Domain Video Processing.

Spatio Temporal Sampling: Sampling for Analog and Digital Video, Sampling Structures for Analog

Video, Two-Dimensional Rectangular Sampling, Two-Dimensional Periodic Sampling, Sampling on 3-D

Structures, Sampling on a Lattice.

Video Modeling: Camera Model-Illumination Model-Object Model-Scene Model-Two-

Dimensional Motion Models.

COURSE OUTCOMES

16CV607-CO1 Understand Video formation principles and Image formation models

16CV607-CO2 Understand algorithms for motion estimation in Video

16CV607-CO3 Understand and apply Block matching algorithms for Motion Estimation

16CV607-CO4 Understand principles in Video compression standards

16CV607-CO5 Understand the principles in Camera Model representation


1. Yao Wang, Jorn Ostermann and Ya-Qin Zhang, “Video Processing and Communications”,

18

Prentice Hall, 2001.

2. A. Murat Tekalp, “Digital Video Processing”, Pearson, 1995

16CV609 DEVELOPMENT TOOLS FOR IMAGE AND VIDEO PROCESSING 0-0-1-1

1. OpenSource Tools for Image Processing and Computer Vision

- OpenCV, OpenGL

2. Beagle Board XM and Beagle Bone Black

- Application Development using OpenCV/Java for Image Processing concepts3. NI Vision System Camera with LabView

- Data Acquisition

- Basic Image Processing concepts4. iRobot

- Matlab programming for Navigation scenario5. LegoMindstrom

- Simple Robot Programming

- Object Detection

COURSE OUTCOMES

16CV609-CO1 To understand the functioning of hardware for running programs for image and video

16CV609-CO2 To develop model with hardware for executing image and video applications

16CV609-CO3 To analyze the different algorithms for computer vision with hardware and software

16CV609-CO4 To develop working prototypes for real time applications.

16MA606 RANDOM PROCESSES AND OPTIMIZATION 4-0-0-4

Probability Theory: Experiments, Outcomes, Probability, conditional probability and Bayes Theorem.

Random Variables and Probability Distributions- Mean and Variance of a Distribution, Binomial, Poisson

and Normal Distributions.

Random Processes: General concepts and definitions – Stationarity in random process,

autocorrelation and properties – Poisson points, Poisson and Gaussian processes. Spectrum

estimation – Ergodicity and mean Ergodic theorem – Power spectral density and properties Markov

processes – Markov Chains – Transition Probability matrix, Classification of states Limiting

Distributions.

Optimization Techniques: Single variable optimization: Optimality criteria – bracketing methods –

region elimination methods – point estimation method – gradient based methods. Multivariable

optimization: Optimality criteria – unidirectional search direct search method gradient based

methods. Lagrangian and Kuhn-Tucker conditions.

1 9

COURSE OUTCOMES

16MAT610-CO1 Understand the basic concepts of random processes and the stationarity.

16MAT610-CO2 Understand the purpose of some special distributions including Markov chains

16MAT610-CO3 Gain knowledge about spectrum estimation and spectral density function

16MAT610-CO4 To understand and apply different types of single variable optimization methods

16MAT610-CO5 To understand and apply different types of multiple variable optimization methods


1. Douglas C. Montgomery and George C. Runger, “Applied Statistics and Probability for

Engineers”, Third Edition, John-Wiley & Sons Inc., 2003.

2. A. Papoulis and Unnikrishna Pillai, “Probability, Random Variables and Stochastic Processes”,Fourth Edition, McGraw Hill, 2002.

3. J Ravichandran, “Probability and Statistics for Engineers”, First Edition, Wiley, 2012.

4. Scott L. Miller, Donald G. Childers, “Probability and Random Processes”, Academic Press,

2012.

5. Kalyanmoy Deb, “Optimization for Engineering Design: Algorithms and Examples”, Prentice

Hall, 2002.

6. Singiresu S. Rao, “Engineering Optimization: Theory and Practice”, Third Edition, New Age

Publishers, 2003.

7. Justin Solomon, “Mathematical Methods for Computer Vision, Robotics, and Graphics”,

Stanford University, 2013.

16CV602 IMAGE ANALYSIS 3-0-1-4

Image Morphology: Binary and gray scale Morphological analysis - Dilation and Erosion - Skeletons and

Object Marking – Granulometry – Morphological Segmentation. Feature extraction: Global image

measurement, feature specific measurement, characterizing shapes, Hough Transform. Representation and

Description: Region Identification – Contour Based and Region Based Shape Representation and

Description – Shape Classes. Flexible shape extraction: active contours, Flexible shape models: active

shape and active appearance. Texture representation and analysis: Statistical Texture Description –

Syntactic Texture Description Methods – Hybrid Texture description Methods – Texture Recognition

Method Applications. Image Understanding: Control Strategies –RANSAC – Point Distribution Models –

Scene Labeling and Constraint Propagation. Image Data Compression: Predictive Compression Methods

– Vector Quantization, DCT, Wavelet, JPEG.

COURSE OUTCOMES

20

16CV602-CO1 Understand and apply image morphological algorithms for image feature extraction

16CV602-CO2 Apply Hough Transform as a tool for shape extraction and feature analysis

16CV602-CO3 Understand and apply Shape descriptors for shape retrieval

16CV602-CO4 Understand and apply Texture analysis algorithms for feature extraction

16CV602-CO5 Understand Image compression techniques for image analysis



Vision”, Third Edition, Cengage Learning, 2007.

2. Tinku Acharya, Ajoy K Ray,“Image Processing- Principles and Applications”, Wiley, 2005.

3. John C. Russ, “The Image Processing Handbook”, Sixth Edition, CRC Press, 2007.

4. Mark S. Nixon, Alberto S. Aguado, “Feature Extraction and Image Processing”, Second Edition,

Academic Press, 2008.

16CV604 PATTERN RECOGNITION AND MACHINE LEARNING 3-0-1-4

Introduction - Pattern recognition systems - The design cycle - Learning and adaptation - Linear modelsfor classification - Discriminant functions (Two and multiple classes) - Least squares classificationfunctions - Fisher’s discriminant analysis for two and multiple classes - Probabilistic generative models -Maximum likelihood solution. Kernel methods: Constructing kernels - Kernel density estimators - Nearestneighbor methods - Gaussian processes and classification - Sparse kernel machines - Support vectormachines - Maximum margin classifiers- Multi-class support vector machine. Graphical models: Bayesian networks - Generative models- Linear Gaussian models - Conditional independence. Mixture models and Expectation maximization: K-means clustering - Mixtures of Gaussian - Expectation maximum for Gaussian mixtures. Continuouslatent variables: Principal component analysis - Applications of principal component analysis - PCA forhigher dimensional data - Factor analysis. Sequential data: Markov models - Hidden Markov models -Maximum likelihood for HMM - Forward-backward algorithm. Combining models - Tree based models -Decision trees - Classification and regression trees (CART).Deep learning for High-level Vision: Introduction to Deep Learning, main types of DeepArchitectures, Application of Deep Learning Architecture to Computer Vision.

COURSE OUTCOMES

16CV604-CO1 Understanding pattern recognition systems and its applications

16CV604-CO2 Analyse statistical decision making for images

16CV604-CO3 Analyze supervised and unsupervised learning approaches for images

16CV604-CO4 Apply dimensionality reduction techniques for images.

16CV604-CO5 Apply deep learning architectures for computer vision

2 1


1. Christopher M. Bishop,“Pattern Recognition and Machine Learning”, Springer, 2006.

2. Richard O. Duda, Peter E. Hart and David G. Stork,“Pattern Classification”, Second Edition,

John Wiley and Sons, 2003.

3. Earl Gose, Richard Johnsonbaugh and Steve Jost,“Pattern Recognition and Image Analysis”,

Prentice Hall of India, 2002.

16CV606 COMPUTER VISION 3-0-1-4

Image Formation: Geometric image formation, Photometric image formation - Camera Models and

Calibration: Camera Projection Models – Orthographic, Affine, Perspective, Projective models. Projective

Geometry, Transformation of 2D and 3D, Internal Parameters, Lens Distortion Models- Local Feature

Detectors and Descriptors: Hessian corner detector, Harris Corner Detector, LOG detector, DOG detector,

SIFT, PCA-SIFT, GLOH, SURF, HOG, Pyramidal HOG, PHOW-Calibration Methods: Linear, Direct,

Indirect and Multiplane methods -

Pose Estimation. Stereo and Multi-view Geometry: Epipolar Geometry, Rectification and Issues related to

Stereo, General Stereo with E Matrix Estimation, Stratification for 2 Cameras, Extensions to Multiple

Cameras, Self-Calibration with Multiple Cameras, 3D reconstruction of cameras and structures, Three

View Geometry.

COURSE OUTCOMES

16CV606-CO1 Understand Image formation and Camera Models

16CV606-CO2 Understand and apply 2D and 3D Transformations

16CV606-CO3 Understand the different techniques for Local feature extraction and tracking

16CV606-CO4 Understand the different category of calibration methods

16CV606-CO5 Understand 3D reconstruction approach for computer vision


1. Forsyth and Ponce, “Computer Vision – A Modern Approach”, Second Edition, Prentice Hall,

2011.

2. Emanuele Trucco and Alessandro Verri, “Introductory Techniques for 3-D Computer Vision”,


3. Olivier Faugeras, “Three Dimensional Computer Vision”, MIT Press, 1993.

4. Richard Szeliski, “Computer Vision: Algorithms and Applications”, Springer, 2011.


Vision”, Third Edition, CL Engineering, 2013.

16CS625 RESEARCH METHODOLOGIES AND SEMINAR 1-0-1-2

22

Note: The course will be evaluated internally by the Department.

This course is intended to be a self study course. Each student can select an area of self study inconsultation with the Faculty, collect and study basic and recent research articles (project reports, reviewarticles, published articles in journals and book chapters.) on the topic. Students will be required to maketwo in -class presentations. The Seminars will be evaluated for grading purpose. The evaluation will bedone by a panel of (at least) two Faculty members.

Topics to be covered:

Selection of project domain; Publication ethics, Tools and evaluation. Selection of tentative project areaand process of literature survey – Literature survey components and procedures Basic components of aresearch paper – procedures and processes, Journal types, Scopus, web of science, Thomson Reuters,Science Citation Index, H-index, Google citations, Presentation of selected project proposal – Oralpresentation. Preparation of a report on the selected project proposal in LaTeX formatAttending special invited lectures, practical orientation in searching and collecting literature throughlibrary, online tools, presenting a seminar on selected project proposal and submitting project reportprepared using LaTeX.

Note: Allotment / Selection of Mentor/Supervisor will be done at Department.

COURSE OUTCOMES

16CS625-CO1 To understand the Meaning, motivation, objectives and type of research

To understand and apply review of literature, identification of variables and16CS625-CO2 construction of hypotheses for formulating a research problem to decide what to

research.

16CS625-CO3 To understand how to conduct a research study through learning system analysis, designof experiment and data collection, for writing a research proposal.

16CS625-CO4 To understand and apply the writing skills for formulating a manuscript.

16CS625-CO5 To understand the Intellectual property rights, Ethics of Research and Plagiarism

16EN600 TECHNICAL WRITING P/F

Technical terms – Definitions – extended definitions – grammar checks – error detection – punctuation –spelling and number rules – tone and style – pre -writing techniques – Online and offline library resources –citing references – plagiarism – Graphical representation – documentation styles – instruction manuals –information brochures – research papers – proposals – reports (dissertation, project reports etc.)

COURSE OUTCOMES

CO1: Understanding of the different styles of various technical bodies

CO2: Knowledge of the various tools available for technical writing

TEXTBOOKS/REFERENCES:

2 3

1. H.L. Hirsch, Essential Communication Strategies for Scientists, Engineers and TechnologyProfessionals, Second Edition, New York: IEEE Press, 2002.

2. P.V. Anderson, Technical Communication: A Reader-Centered Approach, Sixth Edition,Cengage Learning India Pvt. Ltd., New Delhi, 2008, (Reprint 2010).

3. W.Jr. Strunk and E.B.White, The Elements of Style, New York. Alliyan and Bacon, 1999.

16CV798/ 16CV799 DISSERTATION 10 / 12

Each student should select and work on a topic related to his/her field of specialization during summer of

second semester under the supervision of a faculty member. By the end of the third semester he/she must

prepare a report in the approved format and present it. During fourth semester each student should work

further on the topic of the minor project or a new topic under the supervision of a faculty member. By the

end of fourth semester the student has to prepare a report in the approved format and present it.

COURSE OUTCOMES

CO1 Learn to study existing state of the art work in the research area and prepare the literature survey.

CO2 Understand challenges and formulate research problem

CO3 Apply suitable design techniques to develop solutions for the identified problem

CO4 Analyze the results using appropriate evaluation methods

CO5 Prepare report using defined standards adapting professional ethics

16CV701 MODELING AND VISUALISATION 3-0-0-3

Introduction to Graphics, Two-dimensional Geometric Transformations, Three-dimensional Concepts.

Modeling: Three-Dimensional Object Representations: Raw 3D data, Surface Representation, Solid

Representation, High-Level Representation. Reconstruction of 3D Meshes from Polygon Soup: Cell

complex, Solidity Determination, Meshes reconstruction. Advanced Rendering Techniques:Photorealistic

Rendering, Global Illumination, Participating Media Rendering, Ray tracing, Monte Carlo algorithm,

Photon Mapping. Volume Rendering: Volume graphics Overview, Marching cubes, Direct volume

rendering. Surfaces and Meshes. Visualization: Meshes for Visualization, Volume Visualization and

Medical Visualization.

COURSE OUTCOMES

16CV701-CO1 Demonstrate knowledge of the graphics pipeline from 2D images to 3D modelsDisplay solid comprehension of fundamental 3D modeling constructs and the Euclidian

16CV701-CO2 mathematics of 3D spaces.Display fundamental skills in rendering photorealistic 3D scenes with 3D modeling

16CV701-CO3 software.

2 4

Display fundamental skills in designing and realizing Volume rendering for surfaces16CV701-CO4 and Meshes

Demonstrate knowledge of the applying visualization techniques in Life science and16CV701-CO5 medicine.


1. Tomas Akenine Moller, Eric Haines and Naty Hoffman,“ Real-Time Rendering”, Third Edition,

A K Peters Ltd, 2008.

2. Matt Pharr and Greg Humphreys,“Physically Based Rendering: From Theory to

Implementation”, Second Edition, Morgan Kaufmann, 2010.

3. Lars Linsen, Hans Hagen and Bernd Hamann,“Visualization in Medicine and Life Sciences”,

Springer-Verlag Berlin Heidelberg, 2008.

4. Donald Hearn and Pauline Baker, “Computer Graphics”, Second Edition, Prentice Hall of India,

1994.

16CV702 VIDEO ANALYTICS 3-0-0-3

Video Analytics principles. Computer Vision: Challenges - Spatial Domain Processing – Frequency

Domain Processing - Background Modeling - Shadow Detection - Eigen Faces - Object Detection -

Local Features - Mean Shift: Clustering, Tracking - Object Tracking using Active Contours - Action

Recognition - Low Level Image Processing for Action Recognition: Segmentation and Extraction, Local

Binary Pattern, Structure from Motion - Action Representation Approaches: Classification of Various

Dimension of Representation, View Invariant Methods, Gesture Recognition and Analysis, Action

Segmentation.

Case Study: Face Detection and Recognition, Natural Scene Videos, Crowd Analysis, Video

Surveillance, Traffic Monitoring, Intelligent Transport System.

COURSE OUTCOMES

16CV702-CO1 Understand the algorithms available for performing analysis on video data and addressthe challenges

16CV702-CO2 Understand the approaches for identifying and tracking objects and person with motionbased algorithms.

16CV702-CO3 Understand the algorithms available for searching and matching in video content

16CV702-CO4 Analyze approaches for action representation and recognition

16CV702-CO5 Identify, Analyze and apply algorithms for developing solutions for real worldproblems



2. Thierry Bouwmans, Fatih Porikli, Benjamin Höferlin and Antoine Vacavant, “Background

2 5

Modeling and Foreground Detection for Video Surveillance: Traditional and Recent Approaches,

Implementations, Benchmarking and Evaluation", CRC Press, Taylor and Francis Group, 2014.

3. Rastislav Lukac, Boca Raton, FL,“Computational Photography: Methods and Applications, CRC

Press, Taylor and Francis Group, 2010.

4. Ahad, Md. Atiqur Rahman, "Computer Vision and Action Recognition-A Guide for Image

Processing and Computer Vision Community for Action Understanding", Atlantis Press, 2011.

16CV703 ADVANCED COMPUTER VISION 3-0-0-3

Shape from X – Shape from Stereo, Shape from Shading, Shape from Silhouette, Shape from Texture andShape from Focus. Shape Representation: Statistical Shape Models, Active Shape Models, CombinedAppearance Models, Active Appearance Models, View-based Appearance Models, Tracking with View-based Appearance Models. Object Recognition: Shape Correspondence and Shape Matching, PCA, ShapePriors for Recognition, Finding Templates and Recognition, Recognition by Relations betweenTemplates, Robotic vision, Computer Vision on the GPU. Tracking & Video Analysis: Tracking andMotion Understanding - Kalman filters, condensation, particle, Bayesian filters, Hidden Markov models,Change detection and Model-based tracking.

COURSE OUTCOMES

16CV703-CO1 Understand the approaches for Shape Generation

16CV703-CO2 Understand statistical shape model,Active shape Model,Combined Appearance Model,Active Appearance Model and View Based Appearance Model

16CV703-CO3 Understand the algorithms available for Object recognition

16CV703-CO4 Analyze approaches for Tracking and Motion Analysis


1. Forsyth and Ponce, “Computer Vision – A Modern Approach”, Second Edition, Prentice Hall,

2011.

2. Emanuele Trucco and Alessandro Verri, “Introductory Techniques for 3-D Computer Vision”,


3. Olivier Faugeras, “Three Dimensional Computer Vision”, MIT Press, 1993.


5. Richard Hartley and Andrew Zisserman, “Multiple View Geometry in Computer Vision”, Second

Edition, Cambridge University Press, 2004.

16CV704 MACHINE VISION 3-0-0-3

Design and Fabrication of Soft Zoom Lens Applied in Robot Vision - Methods for Reliable Robot Vision

2 6

with a Dioptric System - An Approach for Optimal Design of Robot Vision Systems -Visual Motion

Analysis for 3D Robot Navigation in Dynamic Environments - A Visual Navigation Strategy based on

Inverse Perspective Transformation - Vision-based Navigation Using an Associative Memory. Vision

Based Robotic Navigation: Application to Orthopedic Surgery - Navigation and Control of Mobile

Robot Using Sensor Fusion - Visual Navigation for Mobile Robots - Interactive Object Learning and

Recognition with Multiclass Support Vector Machines - Recognizing Human Gait Types - Environment

Recognition System for Biped Robot Walking Using Vision Based Sensor Fusion - Non Contact 2D and

3D Shape Recognition by Vision System for Robotic Prehension - Image Stabilization in Active Robot

Vision-Real - Time Stereo Vision Applications - Robot vision using 3D TOF systems - Calibration of

Non-SVP Hyperbolic Catadioptric Robotic Vision Systems - Computational Modeling, Visualization,

and Control of 2-D and 3-D Grasping under Rolling Contacts - Towards Real Time Data Reduction and

Feature Abstraction for Robotics Vision - LSCIC Pre Coder for Image and Video Compression - The

Robotic Visual Information Processing System Based on Wavelet Transformation and Photoelectric

Hybrid-Direct Visual Servoing of Planar Manipulators Using Moments of Planar Targets - Industrial

Robot Manipulator Guarding Using Artificial Vision - Remote Robot Vision Control of a Flexible

Manufacturing Cell - Robot Vision in Industrial Assembly and Quality Control Processes - Multi-Task

Active - Vision in Robotics

COURSE OUTCOMES

16CV704-CO1 Understand the fundamental principles behind computer and Machine vision

16CV704-CO2 Understand the state-of-the-art algorithms in machine vision

16CV704-CO3 Apply the machine vision algorithms given an application scenario

Evaluate the existing machine vision algorithms based on their strength, robustness and16CV704-CO4 weakness.


1. Ales Ude, “Robot Vision”, InTech, 2010.

2. Daiki Ito, “Robot Vision: Strategies, Algorithms and Motion Planning”, Nova Science

Publishers Inc, 2009.

3. Christian Wiedemann , Markus Ulrich and Carsten Steger , "Machine Vision Algorithms and

Applications", Wiley-VCH, 2008.

4. E. R. Davies, “Computer and Machine Vision, Theory, Algorithms, Practicalities”, Fourth

Edition, Elsevier, 2012.

5. Ramesh Jain, Rangachar Kasturi and Brian G. Schunck, “Machine Vision”, Tata McGraw-

Hill, 1995.

16CV705 MEDICAL IMAGE ANALYSIS 3-0-0-3

Medical imaging modalities: Planar X-Ray imaging - X-Ray Computed Tomography – Magnetic

Resonance Imaging – Nuclear Imaging – Ultrasonography – Other modalities. Image file formats:

2 7

DICOM and other medical image file formats. Image Enhancement: Fundamental enhancement

techniques- Adaptive Image Filtering – Enhancements by multiscale non linear operators. Image

segmentation: Overview and fundamentals of medical image segmentation – Segmentation using Graph

cuts – Segmentation using fuzzy clustering – Neural networks – Deformable models. Medical image

registration: Rigid body transformation – Non rigid body transformation – Pixel based registration –

Surface based registration – Intensity based registration. Medical image fusion: Linear and non linear

methods – Wavelet based fusion – Pyramidal fusion schemes – Edge preserving fusion algorithms.

Validation of medical image analysis techniques. Case Study: Brain image analysis and atlas construction

– Tumour image analysis and treatment planning – Lung Nodule Analysis - Retinal Image Processing and

analysis - X-Ray image processing and Analysis. Tools: VTK / ITK, Fiji, Mevislab.

COURSE OUTCOMES

16CV705-CO1 Understand the different modalities in Medical Imaging

16CV705-CO2 Understand the state-of-the-art algorithms in Medical Image segmentation

16CV705-CO3 Apply the algorithms for Medical Image Registration

16CV705-CO4 Evaluate the Image Analysis techniques based on their strength, robustness andweakness.


1. Geoff Dougherty, “Medical Image Processing Techniques and Applications”, Springer, 2011.2. Thomas M Deserno, “Biomedical Image Processing”, Springer-Verlag, 2011.

3. A. Ardheshir Goshtasby, “2-D and 3-D Image Registration for Medical, Remote Sensing, and

Industrial Applications”, John Wiley and Sons, 2005.

4. Tania Sthathaki, “Image Fusion: Algorithms and Applications”, Academic Press, 2008.

5. James T . Dobbins III, Sean M . Hames, Bruce H . Hasegawa, Timothy R . DeGrado, James A .

Zagzebski and Richard Frane, “Measurement, Instrumentation, and Sensors Handbook”, Second

Edition: Electromagnetic, Optical, Radiation, Chemical, and Biomedical Measurement, CRC

Press 2014.

16CV706 CONTENT BASED IMAGE AND VIDEO RETRIEVAL 3-0-0-3

Architecture and Design: Introduction - Architecture of content-based image and video retrieval -Designing an image retrieval system - Designing a video retrieval system. Feature extraction andsimilarity measure: Color - Texture - Shape - Spatial relationships - MPEG 7 features. Modeling andanalysis of images: Classification and clustering - Annotation and semantic based retrieval of visual data -Probabilistic models - Relevance feedback. Standards for image data management: Standards relevant toContent based image retrieval - Image compression - Query Specification - Metadata description.Analysis of video: Feature extraction - Semantics understanding - Summarization - Indexing and retrievalof video - Mining large databases. Applications: Architectural and engineering design - Fashion andinterior design - Journalism and advertising -Medical Diagnosis - Geographical Information Systems andRemote Sensing - Education and Training - Web Searching.

2 8

COURSE OUTCOMES

16CV706-CO1 Understand the modules involved in designing CBIVR systems and their applications

16CV706-CO2 Extract different visual features from images and videos

16CV706-CO3 Understand query specification and evaluate the retrieval

16CV706-CO4 Understand indexing and the semantics of visual data

16CV706-CO5 Develop and evaluate visual retrieval algorithms


1. Oge Marques and Borko Furht, “Content Based Image and Video Retrieval”, Multimedia

Systems and Applications, Springer, 2002.

2. Oge Marques, “Practical Image and Video Processing”, Wiley IEEE Press, 2011.

3. Borko Furht and Oge Marques, “Hand Book of Video Databases Design and Applications”,

CRC Press, 2003.

4. Yogita Mistry and Dr. D.T. Ingole, “Survey on Content Based Image Retrieval Systems”,

International Journal of Innovative Research in Computer and Communication Engineering,

2013.

16CV707 DOCUMENT IMAGE ANALYSIS 3-0-0-3

Introduction: Data capture - Document image understanding - Concepts and components. Pixel level

processing: Thresholding – Basic Document Image Binarization – Edge Preserving Binarization

Techniques – Combining Different Binarization Techniques - Case Study: Binarization of Historical

Documents.

Noise reduction and Enhancement: Noise in Conventional and Camera Captured Document Images –

Border Noise and Frame Noise Removal – Circular Noise and Stroke like Pattern Noise Removal –

Clutter Noise Removal – Bleed Through Removal - Case Study: Removing noise from Historical

Documents.

Feature level processing: Introduction - Polygonalization - Critical point detection - Line and curve fitting

- Shape description and recognition - Case Study: Detection of lines, curves and angles from engineering

drawings/maps.

Methodologies: Projection Profile Analysis – Run Length Smearing – Recursive X-Y Cut – White Space

Analysis - Connected component computation - Top Down and Bottom Up Strategies.

Skew detection and correction: Projection Profile based Skew Detection- Nearest Neighbour Clustering

based Skew Detection – Hough Transform based Skew Detection - Slant Estimation.

Layout analysis - Geometric and logic layout analysis - Page decomposition – Region segmentation -

Labelling and classification.

2 9

Text analysis and recognition: Text segmentation – Script, Language and Font identification - Character

segmentation – Machine printed character recognition - Hand written character recognition – Case Study:

Optical Character Recognition (OCR) - Tesseract OCR- Form Processing - Recognition of Braille

characters. Commercial state and future trends.

COURSE OUTCOMES

16CV707-CO1 Understand the different challenges in Binarization techniques

16CV707-CO2 Understand and apply noise reduction and removal for Historical documents

16CV707-CO3 Understand and apply feature level processing for detecting line,curve and angles

16CV707-CO4 Understand the techniques in profile analysis, skew detection and layout analysis

16CV707-CO5 Evaluate the algorithms for Optical Character recognition


1. L. O. Gorman and R. Kasturi, “Document Image Analysis”, IEEE Computer Society Press,1995; IEEE Computer Society Executive Briefings, 1997, ISBN 0-8186-7802-X.

2. David Doermann, Karl Tombre, “Handbook of Document Image Processing and Recognition”,

Springer London, 2014, ISBN 0857298585, 9780857298584.

3. H. Bunke and P. S. P. Wang, “Handbook of Character Recognition and Document Image

Analysis”, World Scientific Press, 1997.

4. H. S. Baird, H. Bunke and K. Yamamoto (Eds.), “Structured Document Image Analysis”,

Springer Verlag, 1992.

5. H. Bunke and Lawrence Spitz (Eds.), “Document Analysis Systems”, Lecture Notes in Computer

Science, Vol. 3872, Springer Verlag, 2006.

16CV708 COMPUTATIONAL INTELLIGENCE FOR IMAGE PROCESSING 3-0-0-3

Computational Intelligence Paradigms overview – Applications of Computational Intelligence for Image

Analysis: Image Enhancement – Thresholding – Segmentation – Image Compression and Reconstruction –

Registration – Image Retrieval – Hyperspectral Image Clustering – Emotion Recognition

COURSE OUTCOMES

16CV708-CO1 To understand the foundations of Computational Intelligence and the properties ofcomputational intelligence.

16CV708-CO2 To understand and apply the structure of biological neural network structure to constructartificial neural network models.

16CV708-CO3 To apply Computational Intelligence for Image Enhancement

16CV708-CO4 To understand the Computational Intelligence for Image Compression andReconstruction

3 0

16CV708-CO5 To understand theory behind Hyper spectral Image clustering


1. Amitava Chatterjee and Patrick Siarry (Eds.), “Computational Intelligence in Image

Processing”, Springer, 2013.

2. Kim-Hui Yap, Ling Guan, Stuart William Perry and Hau San Wong, “Adaptive Image

Processing: A Computational Intelligence Perspective”, Second Edition, Image Processing

Series, CRC Press, 2009.

3. Andries P. Engelbrecht, “Computational Intelligence – An Introduction”, Second Edition,Wiley-Blackwell, 2007.


Vision”, Cengage Learning, 2008.

5. Evangelia Micheli-Tsanakou, “Supervised and Unsupervised Pattern Recognition – Feature

Extraction and Computational Intelligence”, CRC Press, 2000.

16CV709 IMAGE FUSION 3-0-0-3

Introduction to Image Fusion – Spatial and transform domain fusion schemes – fusion rules - Currenttrends in super-resolution image reconstruction- Introduction – Geometric transformation models – Imagedegradation models – State-of-the-art SR methods. Multiresolution analysis – Fundamental principles –wavelet based fusion scheme – pyramid – based fusion scheme – ‘A truos’ wavelet fusion scheme. Imagefusion using ICA. Image fusion using optimization of statistical measurements – Introduction –Mathematical preliminaries – Dispersion Minimisation fusion based methods – Kurtosis MaximisationFusion based methods. Fusion of edge maps using statistical approaches – Introduction – Automatic edgedetection – ROC analysis. Region-based multi focus image fusion – spatial domain fusion – fusion usingsegmented regions. Pixel level image fusion metrics – Signal level performance evolution – Edge basedmetrics – performance of fusion metrics. Objectively adaptive image fusion - forward adaptive – feedbackadaptive schemes – evaluation parameters – Optimal video fusion. Performance evaluation of image fusiontechniques – Signal -to-Noise Ration (SNR) – Peak Signal-to-Noise Ration (PSNR) – Mean Square Error(MSE) – Mutual Information – Fusion Factor – Fusion Symmetry.Case study : Out-of-focus image fusion – Multi-modal image fusion – Image fusion in remote sensingapplications – Image fusion in medical applications.

COURSE OUTCOMES

16CV709-CO1 Understand fundamental principles in image fusion algorithms

16CV709-CO2 Understand and apply multi resolution analysis and wavelet transforms in fusion

16CV709-CO3 Understand the mathematical preliminaries required for image fusion

16CV709-CO4 Understand the existing image fusion evaluation techniques and metrics

16CV709-CO5 Learn and develop image fusion algorithms for the given use cases and evaluate theresults

3 1


1. Tania Stathaki, “Image Fusion- Algorithms and Applications”, First Edition, Academia Press, 2008.

2. Rick S. Blum and Zheng Liu, “Multi-Sensor Image Fusion and Its Applications”, CRC Press, 2005.

16CV710 MULTIMEDIA SECURITY 3-0-0-3

Introduction - Applications and Properties: Applications – Properties – Evaluating Watermarking

Systems. Basic of Cryptographic Techniques: Symmetric Key Encryption – Hash Keys – RSA – DES –

Key Management. Models of Watermarking: Communication based watermarking – Geometric models

of watermarking – Modeling Watermarks Detection by correlation. Watermarking with side information:

Informed Embedding – Informed coding – Dirty paper codes. Perceptual Models: Evaluation –

Perceptual model – Watson’s model – Adaptive watermarking. Robust watermarking – Watermark

Security. Secret Writing and Steganography – Case Study: Watermarking for Copyright Protection.

Biometric Features for User Authentication: Pattern, Speaker and Behavior Recognition - Speaker

Recognition - Face Recognition. Media Sensor Network - Voice over IP (VoIP) Security- Key

Managements and Emerging Technology.

32

COURSE OUTCOMES

16CV710-CO1 To understand the basics of cryptographic techniques

16CV710-CO2 To understand and coding techniques for Watermarking

16CV710-CO3 To Understand and evaluate the perceptual Models

16CV710-CO4 To identify and apply the different mechanisms in security for real time applications


1. Ingemar Cox, Matthew Miller, Jeffrey Bloom and Mathew Miller, “Digital Watermarking: Principles

and Practice”, Morgan Kaufmann Series in Multimedia Information and Systems, 2008.

2. Stefan Katzenbeisser and Fabien A. P. Petitcolas, “Information Hiding Techniques for Steganography

and Digital Watermarking”, Artech House, 1999.

3. Juergen Seitz, “Digital Watermarking for Digital Media”, Information Science Publishing, 2005

4. B. Schneier, “Applied Cryptography”, Second Edition, Wiley, 1996.

5. Menezes, P. van Oorschot and S. Vanstone, “Handbook of Applied Cryptography”, CRC Press, 1997.

16CV711 VISUAL PATTERN ANALYSIS AND MODELING 3-0-0-3

Pattern Analysis and Statistical Learning: Statistical classification, Visual pattern representation, statistical

learning. Unsupervised Learning for Visual Pattern Analysis: Cluster analysis, Clustering algorithms,

Representational Models Component Analysis: Overview, Generative and Discriminative models. Manifold

Learning: global, local and hybrid methods

Functional Approximations: Modeling and Approximating visual data, Lifting schemes, Temporal filtering in

video coding.

Supervised Learning for Visual Classification: Support vector machine, Boosting algorithm. Statistical

Motion Analysis, Tracking of Visual Objects, Robust Visual Tracking, Multi Target Tracking in Video.

Cognition Process: cognitive model, brain research and visual science, visual cognition, cognitive

mechanisms.

COURSE OUTCOMES

16CV711-CO1 Understand visual pattern representational models for objects in images and applyunsupervised learning algorithms

16CV711-CO2 Represent and learn visual patterns in both spatial and temporal domains

16CV711-CO3 Understand statistical visual patterns in images and apply supervised learning algorithms

16CV711-CO4 Understand and analyze visual patterns in video

16CV711-CO5 Understand visual data processing in the cognitive process

16CV711-CO6 Evaluate the statistical visual pattern models for a real world application


3 3

1. Nanning Zheng and Jianru Xue, “Statistical Learning and Pattern Analysis for Image and Video

Processing (Advances in Pattern Recognition Series)”, Springer-Verlag London Limited, 2009.

2. Christopher M Bishop, “Pattern Recognition and Machine Learning” Springer, 2006.

3. Ian T. Babney, “NETLAB: Algorithms for Pattern Recognition (Advances in Pattern Recognition

series)”, Springer, 2002.

16CV712 GPU ARCHITECTURE AND PROGRAMMING 3-0-0-3

Thinking in Parallel: Parallelism Vs. Concurrency, Types and levels of parallelism, Different grains of

parallelism, Flynn’s classification of multi-processors, Introduction to parallelization

and vectorization: Data dependencies, Bernstein conditions for Detection of Parallelism, Motivation forHeterogeneous Computing, Definition of thread and process, Parallel programming models, ParallelProgramming constructs: Synchronization, Deadlocks, Critical sections and Data sharing .

Compute Unified Device Architecture (CUDA): Introduction to heterogeneous architectures-GPU inparticular. Introduction to GPU computing, evolution of GPU pipeline and general purpose computation onGPU, GPU architecture case studies: NVIDIA G80, GT200, Fermi, Kepler etc. CUDA Architecture, CUDAprogramming model, execution model, thread organization: Concept of grid, block and thread, thread indexgeneration, warp. GPU primitives, algorithms and applications: GPU primitives: scan (exclusive or inclusive),scatter, gather, reduce, memory model: Introduction to global, shared, local memories, usage of cache, texturecache, constant memory. CUDA structure, API and library (CUDPP, CUBLAS, FFT etc.) details. CUDAexample programs (Vector dot product, Matrix multiplication (with the usage of tiling and shared memory)etc.). Graph algorithms and dense linear algebra using GPU.GPU Programming using OpenACC:Introductionto OpenACC, basic steps of OpenACC programming. OpenACC programming examples (Image Convolution,Matrix multiplication)

Optimizations and Tools: Memory coalescing; thread and warp divergence, avoiding bank conflicts,Reduction operation using prefix sum example. Usage of shared memory optimally, Performance issues inalgorithms, Need of profilers and analyzers, Introduction to CUDA Tools: GDB, MemCheck, Command line& Visual Profilers, Parallel NSight: Debugger, Analyzer& Graphics Inspector.

COURSE OUTCOMES

16CV712-CO1 Understand GPU computing architecture

16CV712-CO2 Code with GPU programming environments

16CV712-CO3 Design and develop programs that make efficient use of the GPU processing power

16CV712-CO4 Develop solutions to solve computationally intensive problems in various fields


1. David Kirk, Wen-meiHwu , “CUDA: Programming Massively Parallel Processors: A Hands- on

Approach”, Second Edition, Morgan Kaufmann Publishers, 2012.

2. Jason Sanders, Edward Kandrot, “CUDA by Example: An Introduction to General-Purpose GPU

3 4

35

Programming”, First Edition, Addison Wesley, 2010.

3. Michael J. Quinn , “Parallel Programming in C with MPI and OpenMP”, Tata McGraw-Hill, 2003