Course Title: 
Video Analytics
Course Code: 
Year Taught: 
Postgraduate (PG)
School of Engineering

'Video Analytics' is an elective course offered in M. Tech. in Computer Science and Engineering program at School of Engineering, Amrita Vishwa Vidyapeetham.

Introduction: Video Analytics. 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 – Tracking & Video Analysis: Tracking and Motion Understanding – Kalman filters, condensation, particle, Bayesian filters, hidden Markov models, change detection and model based tracking- Motion estimation and Compensation-Block Matching Method, Hierarchical Block Matching, Overlapped Block Motion and compensation,Pel-Recursive Motion Estimation, Mesh Based Method, Optical Flow Method - Motion Segmentation -Thresholding for Change Detection, Estimation of Model parameters - Optical Flow Segmentation-Modified Hough Transform Method- Segmentation for Layered Video Representation-Bayesian Segmentation -Simultaneous Estimation and Segmentation-Motion Field Model - 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.


  1. Richard Szeliski, “Computer Vision: Algorithms and Applications”, Springer, 2011.
  2. Yao Wang, JornOstermann and Ya-Qin Zhang, “Video Processing and Communications”, Prentice Hall, 2001.
  3. A.MuratTekalp, “Digital Video Processing”, Pearson, 1995
  4. Thierry Bouwmans, FatihPorikli, Benjamin Höferlin and Antoine Vacavant, “Background Modeling and Foreground Detection for Video Surveillance: Traditional and Recent Approaches, Implementations, Benchmarking and Evaluation", CRC Press, Taylor and Francis Group, 2014.
  5. Md. Atiqur Rahman Ahad, "Computer Vision and Action Recognition-A Guide for Image Processing and Computer Vision Community for Action Understanding", Atlantis Press, 2011.

Evaluation Pattern

  • Periodical 1 – 15
  • Periodical 2 – 15
  • Continuous Evaluation – 20
  • End Semester – 50

At the end of the course the students will be able to

  Course Outcome Bloom’s Taxonomy Level
CO 1 Understand the algorithms available for performing analysis on video data and address the challenges L2
CO 2 Understand the approaches for identifying and tracking objects and person with motion based algorithms. L2
CO 3 Understand the algorithms available for searching and matching in video content L2
CO 4 Analyze approaches for action representation and recognition L4
CO 5 Identify, Analyze and apply algorithms for developing solutions for real world problems L6