Visualization of Subjects and Their Motion in Video Using Principal Component Analysis

In recent years, videos have become an integral part of daily life and there is an enormous volume of video content. As a result, finding specific videos or scenes within this vast amount of content has become a difficult task. In order to enable quick scene recognition by visualizing both video con...

Full description

Saved in:
Bibliographic Details
Published inInternational journal of mobile computing and multimedia communications Vol. 16; no. 1; pp. 1 - 35
Main Author Sugita, Kaoru
Format Journal Article
LanguageEnglish
Published Hershey IGI Global 17.07.2025
Subjects
Classification
Correlation analysis
Frames (data processing)
Layouts
Methods
Multimedia communications
Principal components analysis
Product reviews
Recognition
Scattering
Video
Visualization
Workloads
Online AccessGet full text

Cover

More Information
Summary:In recent years, videos have become an integral part of daily life and there is an enormous volume of video content. As a result, finding specific videos or scenes within this vast amount of content has become a difficult task. In order to enable quick scene recognition by visualizing both video content and relationships between frames, the authors introduce a method for visualizing the subjects and their motion by arranging video frames in a spatial layout based on their correlations. This method calculates frame correlation matrix by applying Principal Component Analysis (PCA) and arranges the video frames in a spatial layout based on the principal component scores. From the visualization results by scatter plots and 3D space, they found that the scatter plot of minimal frames and 3D space arrangement of video frames were more effective than traditional video playback for scene recognition and workload reduction.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1937-9412
1937-9404
DOI:10.4018/IJMCMC.385390