Abstract
In this work, a new human based video summarization system designed and implemented which receives video frames from surveillance cameras. For analyzing the human centric videos, optical flow used as motion descriptor and histogram of gradients used as feature extractor. During the video motion information of object has been made continuous by adding time information to histograms with gradient information. Template matching method is applied on the continuous gradient information to find beginnings of movement patterns and beginnings of repetition the movement patterns. Summarization is built up on the validation of these extracted patterns with a correlation based search-through subsystem.
Keywords
Automatic summarization
Computer science
Histogram
Computer vision
Artificial intelligence
Extractor
Optical flow
Feature (linguistics)
Feature extraction
Video tracking
Matching (statistics)
Motion (physics)
Pattern recognition (psychology)
Object (grammar)
Image (mathematics)
Mathematics
Subject Areas
Video Analysis and Summarization
·Computer Vision and Pattern Recognition ·Physical Sciences
Human Pose and Action Recognition
·Computer Vision and Pattern Recognition ·Physical Sciences
Human Motion and Animation
·Control and Systems Engineering ·Physical Sciences
Loading…