High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Features Preserving Video Event Detection using Relative Motion Histogram of Bag of Visual Words


Radha Shirbhate , JSPM's BSIOTR,Wagholi,Pune.; Varsha Mahajan, JSPM's BSIOTR,Wagholi,Pune.; Swati Wadghule, JSPM's BSIOTR,Wagholi,Pune.; Arifa Mulani, JSPM's BSIOTR,Wagholi,Pune.


Feature selection, motion relativity and video event detection


Incident discovery in video is a investigate vicinity which attempts to build up a computer system with the capability to robotically read the video and locate the happening from images. Now a day there is a huge demand to find the Event depending on motion relativity and feature selection. In this paper, we propose our predictive based on movement relativity and feature selection for video incident discovery. Furthermost, we recommend a new movement feature, namely The concept of Expanded Relative Motion Histogram of Bag-of-Visual-Words (ERMH-BoW) is used for motion relativity and event detection. In ERMH-BoW, by representing what aspect of an event detection with Bag-of-Visual-Words(BoW), we construct relative motion histograms between different visual words to show the objects activities or how aspect of the event. ERMH-BoW thus integrates both what and how aspects for a absolute event description. Meanwhile, we show that by employing motion relativity, it is invariant to the varying camera movement and able to honestly describe the object activities in an event. We further propose a approach based on information gain and in formativeness weighting to select a cleaner. Our experiments are carried out on several challenging datasets provided by TRECVID for the MED (Multimedia Event Detection) task. It demonstrate that our proposed approach outperforms the state-of-the-art approaches for video event detection.

Other Details

Paper ID: IJSRDV4I20533
Published in: Volume : 4, Issue : 2
Publication Date: 01/05/2016
Page(s): 1009-1012

Article Preview

Download Article