Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences

COLLEGE OF ENGINEERING

UC Berkeley

Can Motion Features Inform Video Aesthetic Preferences?

Scott Chung, Jonathan Sammartino, Jiamin Bai and Brian A. Barsky

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2012-172
June 29, 2012

http://www.eecs.berkeley.edu/Pubs/TechRpts/2012/EECS-2012-172.pdf

We explore a novel approach to evaluate aesthetic qualities of video clips by analyzing key motion features. With a growing number of videos and more interest in sorting through media, evaluating aesthetics is increasingly important. Our approach examines high-level motion-based features in video that account for distribution, trajectory, fluidity, completeness, as well as smooth camera and subject motion. We introduce six fundamental motion features which are derived from basic guidelines in film and then conduct a comprehensive viewer study to examine the correlations between the features and viewer preference.


BibTeX citation:

@techreport{Chung:EECS-2012-172,
    Author = {Chung, Scott and Sammartino, Jonathan and Bai, Jiamin and Barsky, Brian A.},
    Title = {Can Motion Features Inform Video Aesthetic Preferences?},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2012},
    Month = {Jun},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2012/EECS-2012-172.html},
    Number = {UCB/EECS-2012-172},
    Abstract = {We explore a novel approach to evaluate aesthetic qualities of video clips by
analyzing key motion features. With a growing number of videos and more interest
in sorting through media, evaluating aesthetics is increasingly important. Our
approach examines high-level motion-based features in video that account for distribution,
trajectory, fluidity, completeness, as well as smooth camera and subject
motion. We introduce six fundamental motion features which are derived from basic
guidelines in film and then conduct a comprehensive viewer study to examine
the correlations between the features and viewer preference.}
}

EndNote citation:

%0 Report
%A Chung, Scott
%A Sammartino, Jonathan
%A Bai, Jiamin
%A Barsky, Brian A.
%T Can Motion Features Inform Video Aesthetic Preferences?
%I EECS Department, University of California, Berkeley
%D 2012
%8 June 29
%@ UCB/EECS-2012-172
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/2012/EECS-2012-172.html
%F Chung:EECS-2012-172