MU4008 Video Processing and Analytics Syllabus:

MU4008 Video Processing and Analytics Syllabus – Anna University PG Syllabus Regulation 2021

COURSE OBJECTIVES:

 To have a better knowledge about videos representation and its formats
 To know the fundamental concepts of data science and analytics
 To enrich students with video processing for analytics
 To understand the data analytics for processing video content
 To expose the student to emerging trends in video analytics

UNIT I VIDEO FUNDAMENTALS

Basic Concepts and Terminology – Analog Video Standards – Digital Video Basics – Analog-to Digital Conversion – Color Representation and Chroma Sub Sampling – Video Sampling Rate and Standards Conversion – Digital Video Formats –Video Features – Colour, Shape and Textural Features.
Suggested Activities
 In class activity – Numerical problems related to sampling and standards conversion.
 Flipped classroom – Discussion on video features.
Suggested Evaluation Methods
 Online quiz on video features.
 Assignments on sampling and standards conversion.

UNIT II MOTION ESTIMATION AND VIDEO SEGMENTATION

Fundamentals of Motion Estimation – Optical Flow – 2D and 3D Motion Estimation – Block Based Point Correspondences – Gradient Based Intensity Matching – Feature Matching – Frequency Domain Motion Estimation – Video Segmentation.
Suggested Activities
 In-class activity – Numerical problems related to motion estimation.
 External learning – Survey on optical flow techniques.
Suggested Evaluation Methods
 Online quiz on optical flow techniques.
 Assignments on numerical problems in motion estimation.

UNIT III FUNDAMENTAL DATA ANALYSIS

Exploratory Data Analysis – Collection of Data – Graphical Presentation of Data – Classification of Data – Storage and Retrieval of Data – Big Data – Challenges of Conventional Systems – Web Data – Evolution of Analytic Scalability – Analytic Processes and Tools – Analysis vs. Reporting.
Suggested Activities
 In class activity – Graphical presentation of data for visualization.
 External learning – Survey on Modern Data Analytic Tools.
Suggested Evaluation Methods
 Quiz on modern data analytic tools.
 Assignments on data visualization.

UNIT IV MINING DATA STREAMS AND VIDEO ANALYTICS

Introduction To Streams Concepts – Sampling Data in a Stream – Filtering Streams – Counting Distinct Elements in a Stream – Analytic Processes and Tools – Video shot boundary detection – Model Based Annotation and Video Mining – Video Database – Video Categorization – Video Query Categorization.
Suggested Activities
 Flipped classroom on discussion on automatic video trailer generation.
 External learning – Survey on analytic processes and tools.
Suggested Evaluation Methods
 Quiz on video trailer generation.
 Assignments on analytic processes and tools.

UNIT V EMERGING TRENDS

Affective Video Content Analysis – Parsing a Video Into Semantic Segments – Video Indexing and Abstraction for Retrievals – Automatic Video Trailer Generation – Video In painting – Forensic Video Analysis.
Suggested Activities
 External learning – Survey on Affective Video Content Analysis.
 Flipped classroom on discussion on forensic video analysis.
Suggested Evaluation Methods
 Online quiz on forensic video analysis.
 Assignments on affective video content analysis.

PRACTICAL EXERCISES: 30

1. Choose appropriate features for video segmentation for given sample video.
2. Compute two dimension motion estimation using block based match technique.
3. Calculate the motion estimation based on Frequency domain.
4. Compare the video features extracted from a given video dataset using graphical representation.
5. Compute the number of distinct elements found in the given sample data stream.
6. Detect shot boundary for given sample video.
7. Parse the given sample video for indexing and faster retrieval.
8. Generate an automatic video trailer for given sample video.
9. Design simple application using video in painting technique.
10. Mini project for video categorization based on content analysis.

TOTAL: 75 PERIODS

COURSE OUTCOMES:

On completion of the course, the students will be able to:
CO1:Discuss video processing fundamentals
CO2:Analyze video features for segmentation purpose
CO3:Derive numeric problems related to motion estimation
CO4:Process video streams for analytics purpose
CO5:Parse and index video segments
CO6:Design applications for video analytics in current trend

REFERENCES:

1. Roy, A., Dixit, R., Naskar, R., Chakraborty, R.S., “Digital Image Forensics: Theory and Implementation”, Springer, 2018.
2. Paul Kinley, “Data Analytics for Beginners: Basic Guide to Master Data Analytics”, CreateSpace Independent Publishing Platform, 2016.
3. Henrique C. M. Andrade, Bugra Gedik, Deepak S. Turaga, “Fundamentals of Stream Processing: Application Design, Systems, and Analytics”, Cambridge University Press, 2014.
4. Murat Tekalp, “Digital Video Processing” Second Edition, Prentice Hall, 2015.
5. Bart Baesens, “Analytics in a Big Data World: The Essential Guide to Data Science and its Applications“, Wiley, 2014.
6. Oges Marques, “Practical Image and Video Processing Using MATLAB”, Wiley-IEEE Press, 2011.