| |
 |
 |
 |
 |
19-475-0403 REAL-TIME VIDEO ANALYTICS
|
Core/Elective:
Elective Semester: 4 Credits:
4 |
| Course Description |
This course is about video analytics enabling automated analysis of detection of interesting spatial and
temporal events. Image and video analysis include techniques capable of extracting high-level information
from the data. Starting from the foundations of image / video analysis this course covers algorithms applied
in systems for video analytics so as to develop interesting applications including surveillance |
| Course Objectives |
To gain
a working knowledge with imgae and video processing
To understand the analytics on video
To apply the knowledge to develop applications that
use video analytics |
| Course Content |
Module I:
Fundamentals: Image feature extraction: Feature point
detection, Scale Invariant Feature Transform, Edge Detection,
Color features. Multiple View Geometry: Perspective
Projection Camera Model, Epipolar Geometry, Probabilistic
inference, Pattern recognition and Machine learning:
SVM and AdaBoost. Background Modeling and Subtraction:
Kernel Density Approximation, Background Modeling and
Subtraction Algorithms
Module II
Object Detection and Tracking: Pedestrian detection
by boosting local shape features: Tree learning algorithms,
Edgelet features. Occluded pedestrian detection by part
combination. Pedestrian tracking by Associating Detection
Responses. Vehicle Tracking and Recognition: Joint tracking
and Recognition framework, Joint appearance-motion generative
model, Inference algorithm for joint tracking and recognition
Module III
Human Motion Tracking: Image feature representation,
Dimension reduction and Movement dynamics learning.
Human action recognition: Discriminative Gaussian Process
dynamic model. Human Interaction recognition: Learning
human activity, Track-body Synergy framework. Multi-camera
calibration and global trajectory fusion: Non-overlapping
and overlapping cameras. Applications: Attribute-based
people search, Soft biometrics for video surveillance:
Age estimation from face, Gender recognition from face
and body
Module IV
Face Recognition and Gait Analysis: Overview of Recognition
algorithms – Human Recognition using Face, Face Recognition
from still images, Face Recognition from video, Evaluation
of Face Recognition Technologies- Human Recognition
using Gait- HMM Framework for Gait Recognition, View
Invariant Gait Recognition, Role of Shape and Dynamics
in Gait Recognition, Factorial HMM and Parallel HMM
for Gait Recognition, Face Recognition Performance
Module V
Behavioral Analysis and Activity Recognition: Event
Modeling- Behavioral Analysis- Human Activity Recognition-Complex
Activity Recognition- Activity modeling using 3D shape,
Video summarization, Shape based activity models, Suspicious
Activity Detection. Video Segmentation and Key Frame
Extraction: Introduction, Applications of Video Segmentation,
Shot Boundary Detection, Pixel-based Approaches, Block-based
Approaches, Histogram-based Approaches, Clustering-based
Approaches, Performance Measures, Shot Boundary Detection,
Key-frame Extraction
|
| REFERNCES |
1. Francesco
Camastra, Alessandro Vinciarelli, "Machine Learning
for Audio, Image and Video Analysis",Springer Nature,
Second Edition, 2015.
2. Yunqian Ma, Gang Qian, “Intelligent Video Surveillance:
Systems and Technology”, CRC Press, First Edition, 2009.
3. Fredrik Nilsson, Communications Axis, “Intelligent
Network Video: Understanding Modern Video Surveillance
Systems”, CRC Press, Second Edition, 2017.
4. Anthony C. Caputo, “Digital Video Surveillance and
Security”, Butterworth-Heinemann, Second Edition, 2014.
5. Herman Kruegle, “CCTV Surveillance: Video Practices
and Technology”,Butterworth-Heinemann, Second Edition,
2006.
6. Amit K.Roy-Chowdhury, Rama Chellappa, S. Kevin Zhou,
Al Bovik, “Recognition of Humans and Their Activities
Using Video (Synthesis Lectures on Image, Video, and
Multimedia Processing)”, Taxmann Publications Private
Limited, 2005.
7. Richard Szeliski, "Computer Vision: Algorithms and
Applications", Springer, First Edition, 2010.
8. David A. Forsyth, Jean Ponce, "Computer Vision- A
Modern Approach", Pearson Education, Second Edition,
2015. |
|
 |
 |
 |
 |
|
|
|
|
|
|
