Course

Syllabus

Introduction, Spatio-temporal sampling; Sampling structure conversion (without using motion), Motion Analysis, Real versus apparent motion, Spatial-temporal constraint methods (optical flow equation), Block-matching methods, Mesh-based methods, Region-based (parametric) motion modeling, Motion segmentation and layered video representations.
Motion-compensated (MC) filtering, Noise reduction, Signal recovery and general inverse problems, Restoration (deblurring), Super resolution, Mosaicking, Deinterlacing, Frame-rate conversion (MC-Interpolation).
Video Watermarking, Video Compression, Frame-based compression (principles behind MPEG-1/2), Scalable or layered frame-based compression, Object-based compression (principles behind MPEG-4). Surveillance – Video indexing, summarization, and retrieval, Object detection and tracking, Video Analytics.

Text Books / References

1. Wang, J. Ostermann, and Y. Q. Zhang, “Video Processing and Communications”, Prentice    Hall, 1^st edition, 2002. 
2. Tekalp, “Digital Video Processing”, Prentice Hall, 2^nd edition, 2005. 
3. W. Woods, “Multidimensional signal, image and video processing and coding”, Academic Press / Elsevier, 2^nd edition, 2012.
4. Linda G. Shapiro and George C. Stockman, “Computer Vision”, Prentice-Hall Inc., 1^st edition, New Jersey, 2001.

Pre-requisite: Nil

Course Objectives:

• To introduce the fundamentals of video processing, spatio-temporal sampling, and motion analysis techniques.
• To develop an understanding of motion-compensated filtering, video enhancement, and restoration methods.
• To provide insights into video compression, watermarking, video analytics, and surveillance applications

Course Outcomes:

CO-PO Mapping: